BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//ca0b20f33713fd38fb8919aa30267bc2.ias.edu//NONSGML kigkonsult.se i
Calcreator 2.29.14//
CALSCALE:GREGORIAN
UID:8668a27e-e995-461c-bd13-3aa8db426766
X-WR-TIMEZONE:UTC
X-WR-CALNAME:IAS Academic
BEGIN:VEVENT
UID:86c56e43-c94b-4cee-9549-2fb8665dde8c
DTSTAMP:20200602T173219Z
CREATED:20190426T173001Z
DESCRIPTION:
DTSTART:20200108T210000Z
DTEND:20200108T230000Z
LAST-MODIFIED:20200316T172248Z
LOCATION:West Seminar
SUMMARY:NES Seminar
URL:https://www.ias.edu/node/99526
END:VEVENT
BEGIN:VEVENT
UID:169f8970-095c-4cc4-93df-180951661d7c
DTSTAMP:20200602T173219Z
CREATED:20190426T173001Z
DESCRIPTION:
DTSTART:20200115T210000Z
DTEND:20200115T230000Z
LAST-MODIFIED:20200316T172248Z
LOCATION:West Seminar
SUMMARY:NES Seminar
URL:https://www.ias.edu/node/99526
END:VEVENT
BEGIN:VEVENT
UID:376444a4-1608-48f6-843d-a0ad6e0a19af
DTSTAMP:20200602T173219Z
CREATED:20190426T173001Z
DESCRIPTION:
DTSTART:20200122T210000Z
DTEND:20200122T230000Z
LAST-MODIFIED:20200316T172248Z
LOCATION:West Seminar
SUMMARY:NES Seminar
URL:https://www.ias.edu/node/99526
END:VEVENT
BEGIN:VEVENT
UID:86b72720-ae91-4cb8-8acc-dec98d935702
DTSTAMP:20200602T173219Z
CREATED:20190426T173001Z
DESCRIPTION:
DTSTART:20200129T210000Z
DTEND:20200129T230000Z
LAST-MODIFIED:20200316T172248Z
LOCATION:West Seminar
SUMMARY:NES Seminar
URL:https://www.ias.edu/node/99526
END:VEVENT
BEGIN:VEVENT
UID:a8daca7f-5e14-490e-b3b2-17637e7bed83
DTSTAMP:20200602T173219Z
CREATED:20190426T173001Z
DESCRIPTION:
DTSTART:20200205T210000Z
DTEND:20200205T230000Z
LAST-MODIFIED:20200316T172248Z
LOCATION:West Seminar
SUMMARY:NES Seminar
URL:https://www.ias.edu/node/99526
END:VEVENT
BEGIN:VEVENT
UID:2a14a7f5-2ba1-43f8-aab6-a54530b309ab
DTSTAMP:20200602T173219Z
CREATED:20190426T173001Z
DESCRIPTION:
DTSTART:20200212T210000Z
DTEND:20200212T230000Z
LAST-MODIFIED:20200316T172248Z
LOCATION:West Seminar
SUMMARY:NES Seminar
URL:https://www.ias.edu/node/99526
END:VEVENT
BEGIN:VEVENT
UID:8b2223bc-e64c-43b8-a028-d3b6cd043860
DTSTAMP:20200602T173219Z
CREATED:20190426T173001Z
DESCRIPTION:
DTSTART:20200226T210000Z
DTEND:20200226T230000Z
LAST-MODIFIED:20200316T172248Z
LOCATION:West Seminar
SUMMARY:NES Seminar
URL:https://www.ias.edu/node/99526
END:VEVENT
BEGIN:VEVENT
UID:9d7a8879-4632-4c1f-9811-73b5ebc73dd1
DTSTAMP:20200602T173219Z
CREATED:20190426T173001Z
DESCRIPTION:
DTSTART:20200304T210000Z
DTEND:20200304T230000Z
LAST-MODIFIED:20200316T172248Z
LOCATION:West Seminar
SUMMARY:NES Seminar
URL:https://www.ias.edu/node/99526
END:VEVENT
BEGIN:VEVENT
UID:04b01147-017c-4d5d-86fa-8d68b455f435
DTSTAMP:20200602T173219Z
CREATED:20190426T173001Z
DESCRIPTION:
DTSTART:20200311T200000Z
DTEND:20200311T220000Z
LAST-MODIFIED:20200316T172248Z
LOCATION:West Seminar
SUMMARY:NES Seminar
URL:https://www.ias.edu/node/99526
END:VEVENT
BEGIN:VEVENT
UID:71ab828b-3a29-470a-8bd4-8c65ce04a046
DTSTAMP:20200602T173219Z
CREATED:20170413T190001Z
DESCRIPTION:Topic: Lower Bounds in Complexity Theory\, Communication Comple
xity\, and Sunflowers\n\nSpeaker: Toniann Pitassi\, University of Toronto\
; Visiting Professor\, School of Mathematics\n\nVideo: https://video.ias.e
du/members/2020/0302-ToniannPitassi\n\nIn this talk I will discuss the Sun
flower Lemma and similar lemmas that prove (in various contexts) that a se
t/distribution can be partitioned into a structured part and a 'random-loo
king' part. I will introduce communication complexity as a key model for u
nderstanding computation and more generally for reasoning about informatio
n bottlenecks.\n\nI will describe our recent Lifting theorems which prove
in a very general way how efficient communication protocols can be well-ap
proximated by 'simple' protocols. Like Razborov's theorem\, the main compo
nent in the proof is a theorem quite similar to the Sunflower Lemma. We wi
ll show how to use Lifting to reprove Razborov's theorem as well as other
state-of-the-art circuit lower bounds. Time permitting\, we will present o
ther applications of Lifting (such as lower bounds for refuting random CNF
formulas and the resolution of the Alon-Saks-Seymour conjecture) and chal
lenges to proving nonmonotone circuit lower bounds.
DTSTART:20200302T190000Z
DTEND:20200302T200000Z
LAST-MODIFIED:20200401T001044Z
LOCATION:Simonyi Hall 101
SUMMARY:Members' Seminar
URL:https://www.ias.edu/node/73431
END:VEVENT
BEGIN:VEVENT
UID:6e8dedd2-9d40-437f-91e6-2bffa7e27f4b
DTSTAMP:20200602T173219Z
CREATED:20190607T210002Z
DESCRIPTION:Topic: Twisted Calabi-Yau algebras and categories\n\nSpeaker: I
nbar Klang\, Columbia University\n\nThis talk will begin with a discussion
of the string topology category of a manifold M\; this was shown by Cohen
and Ganatra to be equivalent as a Calabi-Yau category to the wrapped Fuka
ya category of T*M. In joint work with Ralph Cohen\, we generalize the Cal
abi-Yau condition from chain complexes to spectra. I'll define our notion
of a twisted Calabi-Yau ring spectrum and discuss examples of interest.
DTSTART:20200302T203000Z
DTEND:20200302T213000Z
LAST-MODIFIED:20200227T193116Z
LOCATION:Princeton University\, Fine 224
SUMMARY:Symplectic Dynamics/Geometry Seminar
URL:https://www.ias.edu/node/100886
END:VEVENT
BEGIN:VEVENT
UID:8056bdd1-0859-4cb9-aa87-2a7d1123a03d
DTSTAMP:20200602T173219Z
CREATED:20190607T213002Z
DESCRIPTION:Topic: Constancy of the dimension for non-smooth spaces with Ri
cci curvature bounded below via regularity of Lagrangian flows\n\nSpeaker:
Elia Bruè\, Scuola Normale Superiore di Pisa\n\nAfter a brief introductio
n to the theory of RCD(K\,N) spaces\, i.e. metric measure spaces with Ricc
i curvature bounded below by K and dimension bounded above by N\, I will p
resent the constancy of the dimension theorem in this setting. This result
generalizes the one obtained by Colding and Naber for Ricci limits. Its p
roof relies on a new regularity result for flow maps of Sobolev velocity f
ields.\n\nThis is based on a joint work with Daniele Semola.
DTSTART:20200302T220000Z
DTEND:20200302T230000Z
LAST-MODIFIED:20200219T210506Z
LOCATION:Simonyi Hall 101
SUMMARY:Analysis Seminar
URL:https://www.ias.edu/node/101451
END:VEVENT
BEGIN:VEVENT
UID:14736e24-0720-4266-b59f-f3a919049527
DTSTAMP:20200602T173219Z
CREATED:20190607T203002Z
DESCRIPTION:Topic: An introduction to Boolean Function Analysis\n\nSpeaker:
Dor Minzer\, Member\, School of Mathematics\n\nVideo: https://video.ias.e
du/csdm/2020/0303-DorMinzer\n\nWe will discuss some of the basic principle
s and results in the study of Boolean-valued functions over the discrete h
ypercube using discrete Fourier analysis. In particular\, we will talk abo
ut basic concepts\, the hypercontractive inequality and the KKL theorem. T
ime permitting\, we will discuss the Fourier-Entropy Conjecture and mentio
n some recent progress towards it.\n\nThe talk is self-contained and no sp
ecial background will be assumed.
DTSTART:20200303T153000Z
DTEND:20200303T173000Z
LAST-MODIFIED:20200401T001126Z
LOCATION:Simonyi Hall 101
SUMMARY:Computer Science/Discrete Mathematics Seminar II
URL:https://www.ias.edu/node/100706
END:VEVENT
BEGIN:VEVENT
UID:fcbf5371-280c-4260-8616-c8ddbc2122fa
DTSTAMP:20200602T173219Z
CREATED:20191106T150001Z
DESCRIPTION:
DTSTART:20200303T170000Z
DTEND:20200303T200000Z
LAST-MODIFIED:20191106T150001Z
LOCATION:Dilworth Room
SUMMARY:Women's International Day
URL:https://www.ias.edu/node/106231
END:VEVENT
BEGIN:VEVENT
UID:3bf49474-0f2a-4719-b367-dcf4eefcf55c
DTSTAMP:20200602T173219Z
CREATED:20190611T180002Z
DESCRIPTION:Topic: What Noisy Convex Quadratics Tell Us about Neural Net Tr
aining\n\nSpeaker: Roger Grosse\, University of Toronto\; Member\, School
of Mathematics\n\nI’ll discuss the Noisy Quadratic Model\, the toy problem
of minimizing a convex quadratic function with noisy gradient observation
s. While the NQM is simple enough to have closed-form dynamics for a varie
ty of optimizers\, it gives a surprising amount of insight into neural net
training phenomena. First\, we’ll look at the problem of adapting learnin
g rates using meta-descent (i.e. differentiating through the training dyna
mics). The NQM illuminates why short-horizon meta-descent objectives vastl
y underestimate the optimal learning rate. Second\, we’ll study how the be
havior of various neural net optimizers depends on the batch size. When is
it beneficial to use preconditioning? Momentum? Parameter averaging? Lear
ning rate schedules? The NQM can generate predictions in seconds which see
m to capture the qualitative behavior of large-scale classification conv n
ets and transformers.
DTSTART:20200303T170000Z
DTEND:20200303T183000Z
LAST-MODIFIED:20200226T153652Z
LOCATION:White-Levy
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/101591
END:VEVENT
BEGIN:VEVENT
UID:c8d0fdcb-55ec-4004-8f11-c974c9297a3a
DTSTAMP:20200602T173219Z
CREATED:20200206T202515Z
DESCRIPTION:
DTSTART:20200303T171500Z
DTEND:20200303T190000Z
LAST-MODIFIED:20200206T202515Z
LOCATION:Princeton University\, Jadwin Hall\, Joseph Henry Room\, Room 102
SUMMARY:Institute for Advanced Study / Princeton University Bahcall Lunch
URL:https://www.ias.edu/node/109081
END:VEVENT
BEGIN:VEVENT
UID:a0d092b6-c6cd-4a1a-af27-c5ff2d2b724c
DTSTAMP:20200602T173219Z
CREATED:20191212T132004Z
DESCRIPTION:
DTSTART:20200304T130000Z
DTEND:20200304T220000Z
LAST-MODIFIED:20191212T132004Z
LOCATION:White Levy
SUMMARY:Workshop on Machine Learning\, Theory and Method in the Social Scie
nces
URL:https://www.ias.edu/node/107096
END:VEVENT
BEGIN:VEVENT
UID:2f7dee12-4d6a-4ef9-9e4f-5f8c4d352ae5
DTSTAMP:20200602T173219Z
CREATED:20191218T164502Z
DESCRIPTION:
DTSTART:20200304T130000Z
DTEND:20200304T220000Z
LAST-MODIFIED:20191218T164502Z
LOCATION:West Lecture
SUMMARY:Workshop on Machine Learning\, Theory and Method in the Social Scie
nces/Alternate
URL:https://www.ias.edu/node/107231
END:VEVENT
BEGIN:VEVENT
UID:23b2b159-6196-4749-8793-42893fa04e09
DTSTAMP:20200602T173219Z
CREATED:20191106T150001Z
DESCRIPTION:
DTSTART:20200304T170000Z
DTEND:20200304T200000Z
LAST-MODIFIED:20191106T150001Z
LOCATION:Dilworth Room
SUMMARY:Women's International Day
URL:https://www.ias.edu/node/106236
END:VEVENT
BEGIN:VEVENT
UID:7152da27-4c21-4f42-8a83-e1d8f6d6f6a1
DTSTAMP:20200602T173219Z
CREATED:20190607T211502Z
DESCRIPTION:Topic: Rationality of algebraic varieties\n\nSpeaker: Alexander
Perry\, Member\, School of Mathematics\n\nI will survey what is known abo
ut the rationality of algebraic varieties\, including recent progress and
open questions. There will be a surprising connection to whiskey.
DTSTART:20200304T230000Z
DTEND:20200305T003000Z
LAST-MODIFIED:20200228T151339Z
LOCATION:Dilworth Room
SUMMARY:Mathematical Conversations
URL:https://www.ias.edu/node/100996
END:VEVENT
BEGIN:VEVENT
UID:3eecbe42-5c44-40a4-8812-9338b94bc436
DTSTAMP:20200602T173219Z
CREATED:20191212T132004Z
DESCRIPTION:
DTSTART:20200305T130000Z
DTEND:20200305T220000Z
LAST-MODIFIED:20191212T132004Z
LOCATION:White Levy
SUMMARY:Workshop on Machine Learning\, Theory and Method in the Social Scie
nces
URL:https://www.ias.edu/node/107101
END:VEVENT
BEGIN:VEVENT
UID:e4ca30ad-4003-4d66-9586-0207ac704f2d
DTSTAMP:20200602T173219Z
CREATED:20191218T164502Z
DESCRIPTION:
DTSTART:20200305T130000Z
DTEND:20200305T163000Z
LAST-MODIFIED:20191218T164502Z
LOCATION:West Lecture
SUMMARY:Workshop on Machine Learning\, Theory and Method in the Social Scie
nces/ Alternate
URL:https://www.ias.edu/node/107236
END:VEVENT
BEGIN:VEVENT
UID:be342447-83de-4063-8866-b0d306dce61b
DTSTAMP:20200602T173219Z
CREATED:20190607T211502Z
DESCRIPTION:
DTSTART:20200305T150000Z
DTEND:20200305T170000Z
LAST-MODIFIED:20191007T174501Z
LOCATION:Simonyi Hall 101
SUMMARY:Working Seminar on Nonabelian Hodge Theory
URL:https://www.ias.edu/node/101106
END:VEVENT
BEGIN:VEVENT
UID:1a6a3fbc-1a60-42ac-b292-1db87033c96e
DTSTAMP:20200602T173219Z
CREATED:20200228T220048Z
DESCRIPTION:Topic: The Origin of Multiphase Galaxy Outflows\n\nSpeaker: Eva
n Schneider\, Princeton University\n\nStar-forming galaxies are often obse
rved to host outflows - gas that is flowing away from the galaxy in phases
ranging from cold molecular clouds to hot X-ray emitting plasma. While th
ese multiphase outflows are routinely observed\, theoretically constrainin
g their origin and evolution has proven difficult. Explaining the prevalen
ce and velocities of the cool ionized phase (T~10^4 K) in particular poses
a challenge. In this talk\, I will discuss a potential dual origin for th
is cool gas. Through a series of extremely high-resolution simulations run
with the GPU-based Cholla code\, I will show that in high star formation
surface density systems\, dense disk gas can be pushed out by the collecti
ve effect of clustered supernovae\, explaining the low-velocity material.
Subsequent shredding and mixing of these clouds creates gas with intermedi
ate densities and temperatures that is prone to radiative cooling\, allowi
ng momentum to transfer between phases and producing high velocity cool ga
s. In addition to explaining the nature of outflows themselves\, these mul
tiphase winds could potentially be a source of the cool photo-ionized gas
that is found in abundance in galaxy halos.
DTSTART:20200305T160000Z
DTEND:20200305T170000Z
LAST-MODIFIED:20200302T142559Z
LOCATION:Bloomberg Hall Astrophysics Library
SUMMARY:Institute for Advanced Study Astrophysics Informal Seminar - Title
Added
URL:https://www.ias.edu/node/109821
END:VEVENT
BEGIN:VEVENT
UID:33de602f-7169-4f88-80bb-68be13020297
DTSTAMP:20200602T173219Z
CREATED:20191106T153002Z
DESCRIPTION:
DTSTART:20200305T170000Z
DTEND:20200305T200000Z
LAST-MODIFIED:20191106T154502Z
LOCATION:West Lecture
SUMMARY:Women's International Day
URL:https://www.ias.edu/node/106241
END:VEVENT
BEGIN:VEVENT
UID:c8e97612-4980-49ee-b906-e53502d79c96
DTSTAMP:20200602T173219Z
CREATED:20191025T210001Z
DESCRIPTION:Topic: Understanding Deep Neural Networks: From Generalization
to Interpretability\n\nSpeaker: Gitta Kutyniok\, Technische Universität Be
rlin\n\nVideo: https://video.ias.edu/machinelearning/2020/0305-GittaKutyni
ok\n\nDeep neural networks have recently seen an impressive comeback with
applications both in the public sector and the sciences. However\, despite
their outstanding success\, a comprehensive theoretical foundation of dee
p neural networks is still missing.\n\nFor deriving a theoretical understa
nding of deep neural networks\, one main goal is to analyze their generali
zation ability\, i.e. their performance on unseen data sets. In case of gr
aph convolutional neural networks\, which are today heavily used\, for ins
tance\, for recommender systems\, already the generalization capability to
signals on graphs unseen in the training set\, typically coined transfera
bility\, has not been thoroughly analyzed yet. As one answer to this quest
ion\, in this talk we will show that spectral graph convolutional neural n
etworks are indeed transferable\, thereby also debunking a common misconce
ption about this type of graph convolutional neural networks.\n\nIf such t
heoretical approaches fail or if one is just given a trained neural networ
k without knowledge of how it was trained\, interpretability approaches be
come necessary. Those aim to 'break open the black box' in the sense of id
entifying those features from the input\, which are most relevant for the
observed output. Aiming to derive a theoretically founded approach to this
problem\, we introduced a novel approach based on rate-distortion theory
coined Rate-Distortion Explanation (RDE)\, which not only provides state-o
f-the-art explanations\, but in addition allows first theoretical insights
into the complexity of such problems. In this talk we will discuss this a
pproach and show that it also gives a precise mathematical meaning to the
previously vague term of relevant parts of the input.
DTSTART:20200305T170000Z
DTEND:20200305T183000Z
LAST-MODIFIED:20200401T001224Z
LOCATION:Dilworth Room
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/105991
END:VEVENT
BEGIN:VEVENT
UID:da61391b-dd01-4adf-8a9a-ba05288a2bb9
DTSTAMP:20200602T173219Z
CREATED:20190607T212006Z
DESCRIPTION:Topic: Euler system\, Eisenstein congruences and the p-adic Lan
glands correspondence\n\nSpeaker: Eric Urban\, Columbia University\n\nI wi
ll discuss how the use of the p-adic Langlands correspondence for GL_2(Q_p
) allows to study Eisenstein congruences of various weight\, level and slo
pes in order to construct Euler systems. I will discuss the GL(2)-case wit
h some details and give some hints how to treat the symplectic case in the
ordinary setting.
DTSTART:20200305T213000Z
DTEND:20200305T223000Z
LAST-MODIFIED:20200225T155302Z
LOCATION:Princeton University\, Fine Hall 214
SUMMARY:Joint IAS/Princeton University Number Theory Seminar
URL:https://www.ias.edu/node/101341
END:VEVENT
BEGIN:VEVENT
UID:4917e24f-76b4-4f38-a341-806b662aa312
DTSTAMP:20200602T173219Z
CREATED:20200128T190748Z
DESCRIPTION:Everyone is welcome. Adam Burrows is the organizer.
DTSTART:20200306T160000Z
DTEND:20200306T170000Z
LAST-MODIFIED:20200128T190748Z
LOCATION:Peyton Hall\, Room 140
SUMMARY:Princeton University Astrophysical Sciences Supernova Discussion Gr
oup
URL:https://www.ias.edu/node/108341
END:VEVENT
BEGIN:VEVENT
UID:80025850-f76b-4f88-b304-50ede94c2bb5
DTSTAMP:20200602T173219Z
CREATED:20200128T185206Z
DESCRIPTION:Topic: General Discussion\n\n
DTSTART:20200306T200000Z
DTEND:20200306T210000Z
LAST-MODIFIED:20200228T164624Z
LOCATION:Peyton Hall\, Room 025
SUMMARY:Princeton University HSC Group Discussion
URL:https://www.ias.edu/node/108271
END:VEVENT
BEGIN:VEVENT
UID:613076d2-2eb8-459e-bfe1-db141d46b68e
DTSTAMP:20200602T173219Z
CREATED:20200128T192045Z
DESCRIPTION:Topic: Wide-Field Survey of Dwarf Satellite Systems Around 10 H
osts in the Local Volume\n\nSpeaker: Discussion Leader Scott Carlsten\, Pr
inceton University\n\nThe discussion paper can be found here: https://arxi
v.org/pdf/1909.07389.pdf.
DTSTART:20200309T150000Z
DTEND:20200309T160000Z
LAST-MODIFIED:20200305T163511Z
LOCATION:Princeton University\, Peyton Hall\, Grand Central\, Room 140
SUMMARY:Princeton University Galread [Galactic/Extragalactic Reading Group]
URL:https://www.ias.edu/node/108426
END:VEVENT
BEGIN:VEVENT
UID:49463528-66cc-4a66-aa4e-3207901a67d2
DTSTAMP:20200602T173219Z
CREATED:20190607T202005Z
DESCRIPTION:Topic: Learning from Censored and Dependent Data\n\nSpeaker: Co
nstantinos Daskalakis \, Massachusetts Institute of Technology\; Member\,
School of Mathematics\n\nVideo: https://video.ias.edu/csdm/2020/0309-Const
antinosDaskalakis\n\nMachine Learning is invaluable for extracting insight
s from large volumes of data. A key assumption enabling many methods\, how
ever\, is having access to training data comprising independent observatio
ns from the entire distribution of relevant data. In practice\, data is co
mmonly missing due to measurement limitations\, legal restrictions\, or da
ta collection and sharing practices. Moreover\, observations are commonly
collected on a network\, a spatial or a temporal domain and may be intrica
tely dependent. Training on data that is censored or dependent is known to
lead to Machine Learning models that are biased.\n\nIn this talk\, we ove
rview recent work on learning from censored and dependent data. We propose
a learning framework which is broadly applicable\, and instantiate this f
ramework to obtain computationally and statistically efficient methods for
linear\, and logistic regression from censored or dependent samples\, in
high dimensions. Our findings are enabled through connections to Statistic
al Physics\, Concentration and Anti-concentration of measure\, and propert
ies of Stochastic Gradient Descent\, and advance some classical challenges
in Statistics and Econometrics. (I will overview works with Dagan\, Dikka
la\, Gouleakis\, Ilyas\, Jayanti\, Kontonis\, Panageas\, Rohatgi\, Tzamos\
, Zampetakis)
DTSTART:20200309T150000Z
DTEND:20200309T160000Z
LAST-MODIFIED:20200401T001326Z
LOCATION:Simonyi Hall 101
SUMMARY:Computer Science/Discrete Mathematics Seminar I
URL:https://www.ias.edu/node/100656
END:VEVENT
BEGIN:VEVENT
UID:fe59cf3b-47cb-443e-af00-f6e914ba9344
DTSTAMP:20200602T173219Z
CREATED:20190802T151502Z
DESCRIPTION:Topic: Dwarf Galaxies as Dark Matter Laboratories\n\nSpeaker: Y
ao-Yuan Mao\, Rutgers University\n\nIn the standard model of cosmology\, d
ark matter can be seen as the building blocks of our universe and is home
to visible galaxies. However\, despite the success of this model and a wid
e array of experiments\, our knowledge about the fundamental nature of dar
k matter remains very limited. Fortunately\, current and upcoming galaxy s
urveys open new windows into fainter dwarf galaxies\, which are excellent
laboratories for studying the nature of dark matter\, as dwarf galaxies ar
e hosted in small dark structures\, and have high dark-to-baryonic mass ra
tio. In this talk\, I will introduce how dwarf galaxies can provide novel
insights into the nature of dark matter\, and how we develop better theore
tical models for the connection between dwarf galaxies and dark matter hal
os. I will also describe the ongoing 'Satellites Around Galactic Analogs'
(SAGA) Survey\, a spectroscopic survey that aims to discover and identify
dwarf galaxy satellite systems around Milky Way-like galaxies outside the
Local Volume\, and how the SAGA dataset and other future surveys will enab
le us to unveil the nature of dark matter.
DTSTART:20200309T163000Z
DTEND:20200309T180000Z
LAST-MODIFIED:20200303T184552Z
LOCATION:Princeton University\, Peyton Hall\, Dome Room\, Room 201
SUMMARY:Institute for Advanced Study/Princeton University Early Universe/Co
smology Lunch Discussion
URL:https://www.ias.edu/node/102626
END:VEVENT
BEGIN:VEVENT
UID:3e87a0bd-cb49-4079-a620-fcd332230fa3
DTSTAMP:20200602T173219Z
CREATED:20170413T190001Z
DESCRIPTION:Topic: Towards a mathematical model of the brain\n\nSpeaker: La
i-Sang Young\, New York University\; Distinguished Visiting Professor\, Sc
hool of Mathematics & Natural Sciences\n\nVideo: https://video.ias.edu/mem
bers/2020/0309-Lai-SangYoung\n\nStriving to make contact with mathematics
and to be consistent with neuroanatomy at the same time\, I propose an ide
alized picture of the cerebral cortex consisting of a hierarchical network
of brain regions each further subdivided into interconnecting layers not
unlike those in artificial neural networks. Each layer is idealized as a 2
D sheet of neurons\, spatially homogeneous with primarily local interactio
ns\, a setup reminiscent of that in statistical mechanics. Zooming into lo
cal circuits\, one gets into the domain of dynamical systems. Here the dyn
amics are characterized by competition and balance between two 'opposing'
groups of agents (Excitatory and Inhibitory neurons)\, trying to negotiate
local dynamic equilibria in response to spatially inhomogeneous external
stimuli. I will illustrate some of these ideas using a biologically realis
tic model of the monkey visual cortex\, built and benchmarked to reproduce
visual phenomena with the ultimate aim of explaining cortical mechanisms.
\n\nThe modeling work discussed is in collaboration with Bob Shapley and L
ogan Chariker.
DTSTART:20200309T180000Z
DTEND:20200309T190000Z
LAST-MODIFIED:20200401T001514Z
LOCATION:Simonyi Hall 101
SUMMARY:Members' Seminar
URL:https://www.ias.edu/node/73436
END:VEVENT
BEGIN:VEVENT
UID:440a36d1-ca4e-48b6-93eb-32246896881f
DTSTAMP:20200602T173219Z
CREATED:20190607T210002Z
DESCRIPTION:Topic: Packing and squeezing Lagrangian tori\n\nSpeaker: Richar
d Hind\, University of Notre Dame\n\nVideo: https://video.ias.edu/sympdyng
eo/2020/0309-RichardHind\n\nWe will ask how many Lagrangian tori\, say wit
h an integral area class\, can be `packed' into a given symplectic manifol
d. Similarly\, given an arrangement of such tori\, like the integral produ
ct tori in Euclidean space\, one can ask about the symplectic size of the
complement. The talk will describe some constructions of balls and Lagrang
ian tori which show the size is larger than expected.\n\nThis is based on
joint work with Ely Kerman.
DTSTART:20200309T193000Z
DTEND:20200309T203000Z
LAST-MODIFIED:20200401T001440Z
LOCATION:Simonyi Hall 101
SUMMARY:Symplectic Dynamics/Geometry Seminar
URL:https://www.ias.edu/node/100891
END:VEVENT
BEGIN:VEVENT
UID:99f86c81-2cca-412e-971a-ba0ec436ab6a
DTSTAMP:20200602T173219Z
CREATED:20190607T213002Z
DESCRIPTION:Topic: Higher order rectifiability and Reifenberg parametrizati
ons\n\nSpeaker: Silvia Ghinassi\, Member\, School of Mathematics\n\nVideo:
https://video.ias.edu/analysis/2020/0309-SilviaGhinassi\n\nWe provide geo
metric sufficient conditions for Reifenberg flat sets of any integer dimen
sion in Euclidean space to be parametrized by a Lipschitz map with Hölder
derivatives. The conditions use a Jones type square function and all state
ments are quantitative in that the Hölder and Lipschitz constants of the p
arametrizations depend on such a function. We use these results to prove s
ufficient conditions for higher order rectifiability of sets and measures.
Key tools for the proof come from Guy David and Tatiana Toro’s parametriz
ation of Reifenberg flat sets in the Hölder and Lipschitz categories. If t
ime allows\, we will discuss some related work in progress and an example
that shows that the conditions are not necessary.
DTSTART:20200309T210000Z
DTEND:20200309T220000Z
LAST-MODIFIED:20200401T001409Z
LOCATION:Simonyi Hall 101
SUMMARY:Analysis Seminar
URL:https://www.ias.edu/node/101456
END:VEVENT
BEGIN:VEVENT
UID:237a4b09-18b1-403e-bef0-f63329365838
DTSTAMP:20200602T173219Z
CREATED:20190607T203002Z
DESCRIPTION:Topic: Introduction to high dimensional expanders \n\nSpeaker:
Irit Dinur\, Weizmann Institute of Science\; Visiting Professor\, School o
f Mathematics\n\nVideo: https://video.ias.edu/csdm/2020/0310-IritDinur\n\n
High dimensional expansion generalizes edge and spectral expansion in grap
hs to hypergraphs (viewed as higher dimensional simplicial complexes). It
is a tool that allows analysis of PCP agreement rests\, mixing of Markov c
hains\, and construction of new error correcting codes. My talk will be de
voted to proving some nice relations between local and global expansion of
these objects.
DTSTART:20200310T143000Z
DTEND:20200310T163000Z
LAST-MODIFIED:20200401T001557Z
LOCATION:Simonyi Hall 101
SUMMARY:Computer Science/Discrete Mathematics Seminar II
URL:https://www.ias.edu/node/100701
END:VEVENT
BEGIN:VEVENT
UID:a6de354f-6a15-419c-9c62-cb3e2908d2bf
DTSTAMP:20200602T173219Z
CREATED:20200310T135432Z
DESCRIPTION:Topic: Your Brain on Energy-Based Models: Applying and Scaling
EBMs to Problems of Interest to the Machine Learning Community Today\n\nSp
eaker: Will Grathwohl\, University of Toronto\n\nVideo: https://video.ias.
edu/machinelearning/2020/0310-WillGrathwohl\n\nIn this talk\, I will discu
ss my two recent works on Energy-Based Models. In the first work\, I discu
ss how we can reinterpret standard classification architectures as class c
onditional energy-based models and train them using recently proposed meth
ods for large-scale EBM training. We find that adding EBM training in this
way provides many benefits while negligibly affecting discriminative perf
ormance\, contrary to other hybrid generative/discriminative modeling appr
oaches. These benefits include improved calibration\, out-of-distribution
detection\, and robustness to adversarial examples.\n\nWhile methods for t
raining EBMs at scale have improved drastically\, they still lag behind ot
her classes of generative models such as flows and GANs. Further\, there h
as been little work on evaluating unnormalized models and comparing them w
ith other model classes. My next work addresses these issues with the Stei
n Discrepancy (SD). The SD is a measure of distance between two distributi
ons that is defined using samples from one distribution and an unnormalize
d model for the other. I explore how the stein discrepancy can be estimate
d in practice at scale and demonstrate applications related to goodness-of
-fit testing and unnormalized model evaluation. Next\, I show how my appro
ach for SD estimation can be turned into a GAN-like training objective for
EBMs which scales to high-dimensional data more gracefully than previous
approaches.
DTSTART:20200310T160000Z
DTEND:20200310T173000Z
LAST-MODIFIED:20200401T001623Z
LOCATION:Dilworth Room
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/110101
END:VEVENT
BEGIN:VEVENT
UID:c0bda292-0573-4e73-a906-754dce24f4df
DTSTAMP:20200602T173219Z
CREATED:20200206T202544Z
DESCRIPTION:
DTSTART:20200310T161500Z
DTEND:20200310T180000Z
LAST-MODIFIED:20200206T202544Z
LOCATION:Princeton University\, Jadwin Hall\, Joseph Henry Room\, Room 102
SUMMARY:Institute for Advanced Study / Princeton University Bahcall Lunch
URL:https://www.ias.edu/node/109086
END:VEVENT
BEGIN:VEVENT
UID:b4b38760-9e61-406f-8657-f97d9c788547
DTSTAMP:20200602T173219Z
CREATED:20200213T181518Z
DESCRIPTION:Topic: Flexible Spectral Methods and High-Level Programming for
PDEs\n\nSpeaker: Keaton Burns\, Massachusetts Institute of Technology (MI
T)\n\nThe large-scale numerical solution of PDEs is an essential part of s
cientific research. Decades of work have been put into developing fast num
erical schemes for specific equations\, but computational research in many
fields is still largely software-limited. Here I will discuss how algorit
hmic flexibility and composability can enable new science\, as illustrated
by the Dedalus Project. Dedalus is an open-source Python framework that a
utomates the solution of general PDEs using spectral methods. High-level a
bstractions allow users to symbolically specify equations\, parallelize an
d scale their solvers to thousands of cores\, and perform arbitrary analys
is with the computed solutions. These features are enabling us to perform
novel simulations of astrophysical and geophysical fluids with modern math
ematical techniques. I will discuss applications using new bases for tenso
r-valued equations in spherical domains\, immersed boundary methods for mu
ltiphase flows\, and multi-domain simulations interfacing Dedalus with oth
er PDE and integral equation solvers.
DTSTART:20200311T150000Z
DTEND:20200311T160000Z
LAST-MODIFIED:20200304T203508Z
LOCATION:407 Jadwin Hall\, 4th Floor\, PCTS Seminar Room
SUMMARY:Princeton University Program in Applied and Computational Mathemati
cs (PACM) Special Colloquium
URL:https://www.ias.edu/node/109241
END:VEVENT
BEGIN:VEVENT
UID:a065445f-4186-4ebb-b9cc-02fb15e38a8e
DTSTAMP:20200602T173219Z
CREATED:20190514T144502Z
DESCRIPTION:**In order to help mitigate the growing risk of COVID-19 transm
ission\, this film is postponed.\n\nWe will send out an announcement when
the event is rescheduled.**
DTSTART:20200311T200000Z
DTEND:20200311T223000Z
LAST-MODIFIED:20200309T201501Z
LOCATION:Wolfensohn Hall
SUMMARY:Institute Film Series: Dark Money
URL:https://www.ias.edu/node/100056
END:VEVENT
BEGIN:VEVENT
UID:56303465-434c-46d5-a6a8-f8d1e44920a0
DTSTAMP:20200602T173219Z
CREATED:20200310T151124Z
DESCRIPTION:Topic: Improved Bounds on Minimax Regret under Logarithmic Loss
via Self-Concordance\n\nSpeaker: Blair Bilodaeu\, University of Toronto\n
\nWe study sequential probabilistic prediction on data sequences which are
not i.i.d.\, and even potentially generated by an adversary. At each roun
d\, the player assigns a probability distribution to possible outcomes and
incurs the log-likelihood of the observed outcome under said distribution
. Without assumptions on the data-generating process\, one cannot control
the total loss\, and so it is standard to study the regret\, i.e.\, the di
fference between the player’s total loss and the total loss of the single
best forecaster in some class of “experts”. We present a novel approach to
the open problem of bounding the minimax regret under logarithmic loss fo
r arbitrarily large expert classes by exploiting the self-concordance prop
erty of logarithmic loss. Our regret bound depends on the metric entropy o
f the expert class and matches previous best known results for arbitrary e
xpert classes. We observe a polynomial improvement in dependence on the ti
me horizon for high-dimensional classes.
DTSTART:20200311T200000Z
DTEND:20200311T213000Z
LAST-MODIFIED:20200310T151124Z
LOCATION:Simonyi 101
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/110106
END:VEVENT
BEGIN:VEVENT
UID:d74b3e92-6865-4c95-bea0-8822ce2d6402
DTSTAMP:20200602T173219Z
CREATED:20190607T211502Z
DESCRIPTION:
DTSTART:20200312T140000Z
DTEND:20200312T160000Z
LAST-MODIFIED:20191007T174501Z
LOCATION:Simonyi Hall 101
SUMMARY:Working Seminar on Nonabelian Hodge Theory
URL:https://www.ias.edu/node/101111
END:VEVENT
BEGIN:VEVENT
UID:30730c22-0f68-4524-96e4-819e067940d5
DTSTAMP:20200602T173219Z
CREATED:20200311T192547Z
DESCRIPTION:Topic: An explicit realization of stable L-packets on (split) c
lassical groups\n\nSpeaker: David Soudry\, Tel Aviv University\n\n**Please
note: All in-person IAS seminars have been cancelled. Seminar will be liv
estreamed. Please follow link: https://youtu.be/vZz1SFWHiog\n\nRecently\,
Cai\, Friedberg\, Ginzburg and Kaplan generalized the doubling method of P
iatetski-Shapiro and Rallis. They found global integrals\, which represent
the standard L-function for pairs of irreducible\, automorphic\, cuspidal
representations \pi - on a (split) classical group G\, and \tau - on GL(n
). The representation \pi need not have any particular model (such as a Wh
ittaker model\, or Bessel model\, etc.). These integrals suggest an explic
it descent map (an inverse to Langlands functorial lift) from GL(n) to G (
appropriate G). I will show that a certain Fourier coefficient applied to
a residual Eisenstein series\, induced from a Speh representation\, corres
ponding to a self-dual \tau\, is equal to the direct sum of irreducible cu
spidal representations \sigma \otimes \sigma'\, on G x G \, where \sigma r
uns over all irreducible cuspidal representations\, which lift to \tau (\s
igma' is the complex conjugate of an outer conjugation of \sigma). This is
a joint work with David Ginzburg.
DTSTART:20200312T203000Z
DTEND:20200312T213000Z
LAST-MODIFIED:20200312T171540Z
LOCATION:Seminar cancelled. Event to be livestreamed. Link posted by 3:30P
M
SUMMARY:Joint IAS/Princeton University Number Theory Seminar
URL:https://www.ias.edu/node/110111
END:VEVENT
BEGIN:VEVENT
UID:4c224834-caa4-4a14-9460-2142993d8089
DTSTAMP:20200602T173219Z
CREATED:20200128T190816Z
DESCRIPTION:Everyone is welcome. Adam Burrows is the organizer.
DTSTART:20200313T150000Z
DTEND:20200313T160000Z
LAST-MODIFIED:20200128T190816Z
LOCATION:Peyton Hall\, Room 140
SUMMARY:Princeton University Astrophysical Sciences Supernova Discussion Gr
oup
URL:https://www.ias.edu/node/108346
END:VEVENT
BEGIN:VEVENT
UID:15bf5ec7-d891-4818-aa9a-8dd41e940db2
DTSTAMP:20200602T173219Z
CREATED:20200213T182551Z
DESCRIPTION:Topic: Computational Challenges in Black Hole Astrophysics\n\nS
peaker: Charles Gammie\, University of Illinois\, Urbana-Champaign)\n\nRec
ent astronomical imaging of a black hole has focused attention on problems
related to the production of observable radiation from plasma flow close
to a black hole. Here I will describe the most commonly used computational
scheme for integration of general relativistic magnetohydrodynamic (GRMHD
) models for the plasma\, highlighting successes\, future directions\, and
limitations. The limitations motivate the development of a new\, paramete
rized statistical model that can be matched to GRMHD evolution for particu
lar values of the parameters and can be used to easily generate synthetic
data for all values of the parameters. This opens the door to novel observ
ational tests of GRMHD models. Our statistical model is expected to find a
pplications in simulating turbulent flows near black holes and elsewhere i
n astrophysics.
DTSTART:20200313T160000Z
DTEND:20200313T170000Z
LAST-MODIFIED:20200309T154530Z
LOCATION:Robertson Hall\, Room 001
SUMMARY:Princeton University Program in Applied and Computational Mathemati
cs (PACM) Special Colloquium - Starts 12 pm
URL:https://www.ias.edu/node/109246
END:VEVENT
BEGIN:VEVENT
UID:55fa78a8-a064-4682-b53c-411012634713
DTSTAMP:20200602T173219Z
CREATED:20200128T185227Z
DESCRIPTION:Topic: General Discussion\n\n
DTSTART:20200313T190000Z
DTEND:20200313T200000Z
LAST-MODIFIED:20200228T164541Z
LOCATION:Peyton Hall\, Room 025
SUMMARY:Princeton University HSC Group Discussion
URL:https://www.ias.edu/node/108276
END:VEVENT
BEGIN:VEVENT
UID:7b7ff8ff-f27e-4f45-981c-a8b801124071
DTSTAMP:20200602T173219Z
CREATED:20200313T194133Z
DESCRIPTION:Topic: Feature purification: How adversarial training can perfo
rm robust deep learning\n\nSpeaker: Yuanzhi Li\, Carnegie Mellon Universit
y\n\nVideo: https://video.ias.edu/csdm/2020/0316-YuanzhiLi\n\nTo connect t
o the CSDM Seminar via Zoom\, please do the following:\n1 - If you have Zo
om installed on your device\, enter the following meeting ID: 360043913 \,
click Join Meeting.\n2 - If you do not have Zoom installed\, click the fo
llowing link to download and install: https://theias.zoom.us/j/360043913\n
3 - Once installed\, click the very same link to connect to the meeting.\n
\nWhy deep learning models\, trained on many machine learning tasks\, can
obtain nearly perfect predictions of unseen data sampled from the same dis
tribution but are extremely vulnerable to small perturbations of the input
? How can adversarial training improve the robustness of the neural networ
ks over such perturbations? In this work\, we developed a new principle ca
lled 'feature purification''. Mathematically\, the principle states that f
or a multi-layer ReLU neural network\, let w_i^0 be the weight of the i-th
neuron at initialization\, w_i be its weight after clean training (starti
ng from w_i^0)\, and w_i' as its weight after adversarial training (starti
ng from w_i)\, then for most of the neurons\, there are scaling factors \b
eta_i and vectors v_i such that w_i = \beta_i w_i' + v_i with ||\beta_i w_
i' || > ||v_i||\, while both w_i and w_i' have nearly zero correlations wi
th w_i^0.\n\nConceptually\, the principle says that while both clean train
ing and adversarial training will discover certain features fundamentally
different from the initial values\, clean training actually already discov
ers a big portion of the 'robust features' obtained by adversarial trainin
g. Thus\, instead of needing to learn new 'robust features' or completely
remove 'non-robust features'\, adversarial training can simply robustify t
he neural network by purifying the clean trained features.\n\nIn this work
\, we present both experiments demonstrating the principle\, and a mathema
tical model formally proving it. In particular\, we show that for certain
binary classification tasks\, when we train a two-layer ReLU network using
SGD\, (1). Both clean training and adversarial training will learn featur
es with nearly zero correlations with the (random) initialization. (2). Af
ter clean training\, the network will provably achieve > 99% clean accurac
y but 99% robust accuracy in polynomial samples and running time by simply
'purifying' the clean trained features. Our lower bound that clean traine
d network has\n\nOur work also sheds light on why 'adversary examples are
not bugs': They are not because neural networks are over-fitting to the da
ta set due to the high complexity of the models with insufficiently many t
raining data\, but rather are an intrinsic property of the gradient-descen
t type training algorithms.
DTSTART:20200316T150000Z
DTEND:20200316T160000Z
LAST-MODIFIED:20200401T001742Z
LOCATION:https://theias.zoom.us/j/360043913
SUMMARY:Computer Science/Discrete Mathematics Seminar I
URL:https://www.ias.edu/node/110126
END:VEVENT
BEGIN:VEVENT
UID:89c6c1e8-5e20-4ab2-82e4-054f79b2fc6c
DTSTAMP:20200602T173219Z
CREATED:20200316T134918Z
DESCRIPTION:Topic: Physical Correlations of the Scatter between Galaxy Mass
\, Stellar Content\, and Halo Mass\n\nSpeaker: Discussion Leader Song Huan
g\, Princeton University\n\nThe discussion paper can be found here: https:
//ui.adsabs.harvard.edu/abs/2020MNRAS.tmp...82B/abstract.
DTSTART:20200316T150000Z
DTEND:20200316T160000Z
LAST-MODIFIED:20200330T174732Z
LOCATION:Remote Access Only
SUMMARY:Princeton University Galread [Galactic/Extragalactic Reading Group]
- Added
URL:https://www.ias.edu/node/110131
END:VEVENT
BEGIN:VEVENT
UID:efc09d8c-16e5-4ea7-8956-098ea6a80553
DTSTAMP:20200602T173219Z
CREATED:20190416T134501Z
DESCRIPTION:Topic: Effects of Overlapping Sources in Cosmic Shear Statistic
al Noise\n\nSpeaker: Javier Sànchez\, Fermilab\n\n
DTSTART:20200316T163000Z
DTEND:20200316T180000Z
LAST-MODIFIED:20200330T174455Z
LOCATION:Remote Access Only
SUMMARY:Institute for Advanced Study/Princeton University Early Universe/Co
smology Lunch Discussion
URL:https://www.ias.edu/node/99146
END:VEVENT
BEGIN:VEVENT
UID:f22876cf-f383-4ea4-9473-d8cde01684ec
DTSTAMP:20200602T173219Z
CREATED:20190607T203002Z
DESCRIPTION:Topic: Sharp Thresholds and Extremal Combinatorics\n\nSpeaker:
Dor Minzer\, Member\, Institute for Advanced Study\n\nVideo: https://video
.ias.edu/csdm/2020/0317-DorMinzer\n\nTo connect to the CSDM Seminar via Zo
om\, please do the following:\n1 - If you have Zoom installed on your devi
ce\, enter the following meeting ID: 360043913 \, click Join Meeting.\n2 -
If you do not have Zoom installed\, click the following link to download
and install: https://theias.zoom.us/j/360043913\n3 - Once installed\, clic
k the very same link to connect to the meeting.\n\nConsider the p-biased d
istribution over ${0\,1}^n$\, in which each coordinate independently is sa
mpled according to a $p$-biased bit. A sharp-threshold result studies the
behavior of Boolean functions over the hypercube under different p-biased
measures\, and in particular whether the function experiences a phase tran
sition between two\, close p's. While the theory of sharp-thresholds is we
ll understood for p's that are bounded away from 0 and 1\, it is much less
so for values of p that are close to 0 or 1.\n\nIn this talk\, we will fi
rst discuss classical sharp-threshold results\, and demonstrate an applica
tion of them in Extremal Combinatorics [Dinur-Friedgut]. We will then disc
uss newer sharp-threshold results. Time permitting\, we will mention appli
cations to two problems in extremal combinatorics: the Erdos matching conj
ecture\, and the problem of determining the largest family of vectors in [
m]^n that avoids a fixed\, constant-sized intersections.\n\nBased on joint
works with Peter Keevash\, Noam Lifshitz and Eoin Long.
DTSTART:20200317T143000Z
DTEND:20200317T163000Z
LAST-MODIFIED:20200401T001815Z
LOCATION:https://theias.zoom.us/j/360043913
SUMMARY:Computer Science/Discrete Mathematics Seminar II
URL:https://www.ias.edu/node/100696
END:VEVENT
BEGIN:VEVENT
UID:58f4b0de-ad9d-4457-8aa6-3ceb9b654bb8
DTSTAMP:20200602T173219Z
CREATED:20191203T183001Z
DESCRIPTION:Topic: *Cancelled\n\n
DTSTART:20200317T160000Z
DTEND:20200317T173000Z
LAST-MODIFIED:20200316T174943Z
LOCATION:
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/106931
END:VEVENT
BEGIN:VEVENT
UID:ed104885-5569-44ff-8e90-638bf1914a99
DTSTAMP:20200602T173219Z
CREATED:20190607T211502Z
DESCRIPTION:
DTSTART:20200319T140000Z
DTEND:20200319T160000Z
LAST-MODIFIED:20200316T134755Z
LOCATION:
SUMMARY:Working Seminar on Nonabelian Hodge Theory
URL:https://www.ias.edu/node/101116
END:VEVENT
BEGIN:VEVENT
UID:a75a37b7-607b-48a0-9f03-2e6265cd60f7
DTSTAMP:20200602T173219Z
CREATED:20191025T210001Z
DESCRIPTION:Topic: *Cancelled\n\n
DTSTART:20200319T160000Z
DTEND:20200319T173000Z
LAST-MODIFIED:20200316T175419Z
LOCATION:
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/106001
END:VEVENT
BEGIN:VEVENT
UID:98a24e1d-85b0-4d48-ab5e-409de30b62a6
DTSTAMP:20200602T173219Z
CREATED:20200204T185406Z
DESCRIPTION:Topic: *Cancelled\n\n
DTSTART:20200320T190000Z
DTEND:20200320T200000Z
LAST-MODIFIED:20200316T175347Z
LOCATION:Simonyi 101
SUMMARY:Analysis - Mathematical Physics
URL:https://www.ias.edu/node/108881
END:VEVENT
BEGIN:VEVENT
UID:f9f8be24-9e4f-45da-9c61-80525c8f1762
DTSTAMP:20200602T173219Z
CREATED:20200128T185246Z
DESCRIPTION:Topic: General Discussion\n\n
DTSTART:20200320T190000Z
DTEND:20200320T200000Z
LAST-MODIFIED:20200313T203330Z
LOCATION:To Be Announced
SUMMARY:Princeton University HSC Group Discussion
URL:https://www.ias.edu/node/108281
END:VEVENT
BEGIN:VEVENT
UID:4df50ac9-fe0d-439f-a9ea-75404f55e379
DTSTAMP:20200602T173219Z
CREATED:20200204T185902Z
DESCRIPTION:Topic: *Cancelled\n\n
DTSTART:20200320T210000Z
DTEND:20200320T220000Z
LAST-MODIFIED:20200316T175516Z
LOCATION:Simonyi 101
SUMMARY:Analysis - Mathematical Physics
URL:https://www.ias.edu/node/108886
END:VEVENT
BEGIN:VEVENT
UID:1ac255d0-c575-48b0-b1fe-751ee0f288c5
DTSTAMP:20200602T173219Z
CREATED:20191127T182004Z
DESCRIPTION:Topic: *Cancelled\n\n
DTSTART:20200323T120000Z
DTEND:20200323T210000Z
LAST-MODIFIED:20200319T134008Z
LOCATION:West Lecture
SUMMARY:Mathematics Seminars
URL:https://www.ias.edu/node/106796
END:VEVENT
BEGIN:VEVENT
UID:9b65bad5-9385-4203-ac2e-7f9e1c76a06a
DTSTAMP:20200602T173219Z
CREATED:20190607T202005Z
DESCRIPTION:Topic: Optimal tiling the Euclidean space using symmetric bodie
s\n\nSpeaker: Mark Braverman\, Princeton University\n\nWe say a body B til
es the space R^n if the shifted bodies (B+z)\, for z\in Z^n\, are all disj
oint and cover R^n.\n\nIn this talk\, we consider the problem of finding t
he least surface area a tiling body B can have\, for bodies B symmetric un
der coordinate permutation. This problem arises naturally in the study of
parallel repetition theorems\, which are an important component in many ha
rdness of approximation results. For general bodies\, the isoperimeric ine
quality implies that any tiling body B must have surface area at least \sq
rt{n}\, and somewhat surprisingly this bound is asymptotically tight.\n\nW
ith the symmetry requirement\, the trivial upper bound is O(n) for the uni
t cube\, and the trivial lower bound is still \sqrt{n}. In this talk we sh
ow that the answer for the symmetric case is \Theta(n/\sqrt{\log n}).\n\nW
e will discuss connections to the parallel repetition theorem. Specificall
y\, while the strong parallel repetition conjecture was refuted by Raz in
2008\, our result suggests that there might be important special cases whe
re it still applies.\n\nJoint work with Dor Minzer
DTSTART:20200323T150000Z
DTEND:20200323T160000Z
LAST-MODIFIED:20200323T105516Z
LOCATION:https://theias.zoom.us/j/360043913
SUMMARY:Computer Science/Discrete Mathematics Seminar I
URL:https://www.ias.edu/node/100666
END:VEVENT
BEGIN:VEVENT
UID:86c5f09d-b976-4a23-b77d-4defe485e2ad
DTSTAMP:20200602T173219Z
CREATED:20200320T182655Z
DESCRIPTION:Topic: Early Dark Energy Does Not Restore Cosmological Concorda
nce\n\nSpeaker: Discussion Leader Colin Hill\, Columbia University\n\n
DTSTART:20200323T163000Z
DTEND:20200323T180000Z
LAST-MODIFIED:20200330T174550Z
LOCATION:Remote Access Only
SUMMARY:Institute for Advanced Study/Princeton University Early Universe/Co
smology Lunch Discussion - Speaker Added
URL:https://www.ias.edu/node/110191
END:VEVENT
BEGIN:VEVENT
UID:cee16c73-36e6-4323-bb54-35659308ff3a
DTSTAMP:20200602T173219Z
CREATED:20190607T210002Z
DESCRIPTION:Topic: *Cancelled\n\n
DTSTART:20200323T193000Z
DTEND:20200323T203000Z
LAST-MODIFIED:20200316T175934Z
LOCATION:Simonyi Hall 101
SUMMARY:Symplectic Dynamics/Geometry Seminar
URL:https://www.ias.edu/node/100901
END:VEVENT
BEGIN:VEVENT
UID:e8428270-a8e8-4073-b976-effb8da951da
DTSTAMP:20200602T173219Z
CREATED:20190607T214502Z
DESCRIPTION:Topic: *Cancelled\n\n
DTSTART:20200323T210000Z
DTEND:20200323T220000Z
LAST-MODIFIED:20200316T180024Z
LOCATION:Simonyi Hall 101
SUMMARY:Analysis Seminar
URL:https://www.ias.edu/node/101466
END:VEVENT
BEGIN:VEVENT
UID:ce22389a-fe7b-417f-bbff-a56276fa6a3d
DTSTAMP:20200602T173219Z
CREATED:20190607T203002Z
DESCRIPTION:Topic: High dimensional expansion and agreement testing\n\nSpea
ker: Irit Dinur\, Weizmann Institute of Science\; Visiting Professor\, Sch
ool of Mathematics\n\nVideo: https://video.ias.edu/csdm/2020/0324-IritDinu
r\n\nIn this talk I will describe the notion of 'agreement tests' that are
motivated by PCPs but stand alone as a combinatorial property-testing que
stion. I will show that high dimensional expanders support agreement tests
\, thereby derandomizing direct product tests in a very strong way.
DTSTART:20200324T143000Z
DTEND:20200324T163000Z
LAST-MODIFIED:20200401T001920Z
LOCATION:https://theias.zoom.us/j/360043913
SUMMARY:Computer Science/Discrete Mathematics Seminar II
URL:https://www.ias.edu/node/100691
END:VEVENT
BEGIN:VEVENT
UID:239cc37e-f786-4892-adfa-f6e5854a2094
DTSTAMP:20200602T173219Z
CREATED:20200219T154157Z
DESCRIPTION:Topic: *Cancelled\n\n
DTSTART:20200325T220000Z
DTEND:20200325T233000Z
LAST-MODIFIED:20200319T133854Z
LOCATION:
SUMMARY:Mathematical Conversations
URL:https://www.ias.edu/node/109436
END:VEVENT
BEGIN:VEVENT
UID:42f41460-4a3e-4f09-a68c-70401ff85828
DTSTAMP:20200602T173219Z
CREATED:20190607T211502Z
DESCRIPTION:Topic: *Cancelled\n\n
DTSTART:20200326T140000Z
DTEND:20200326T160000Z
LAST-MODIFIED:20200319T133931Z
LOCATION:Simonyi Hall 101
SUMMARY:Working Seminar on Nonabelian Hodge Theory
URL:https://www.ias.edu/node/101121
END:VEVENT
BEGIN:VEVENT
UID:82f9c147-53ec-40d1-9355-81a5e1145b5d
DTSTAMP:20200602T173219Z
CREATED:20191025T210001Z
DESCRIPTION:Topic: Margins\, perceptrons\, and deep networks\n\nSpeaker: Ma
tus Telgarsky\, University of Illinois\n\nVideo: https://video.ias.edu/mac
hinelearning/2020/0326-MatusTelgarsky\n\nThis talk surveys the role of mar
gins in the analysis of deep networks. As a concrete highlight\, it sketch
es a perceptron-based analysis establishing that shallow ReLU networks can
achieve small test error even when they are quite narrow\, sometimes even
logarithmic in the sample size and inverse target error. The analysis and
bounds depend on a certain nonlinear margin quantity due to Nitanda and S
uzuki\, and can lead to tight upper and lower sample complexity bounds.\n
\nJoint work with Ziwei Ji.
DTSTART:20200326T160000Z
DTEND:20200326T173000Z
LAST-MODIFIED:20200401T001950Z
LOCATION:https://illinois.zoom.us/j/741628827
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/106006
END:VEVENT
BEGIN:VEVENT
UID:d7e11121-ff5b-4be6-994b-49e422b2c3f8
DTSTAMP:20200602T173219Z
CREATED:20200323T135723Z
DESCRIPTION:Topic: Using Quantum Mechanics to Study the Gravitational Three
-body Problem for Special Resonance Motions\n\nSpeaker: Ed Belbruno\, Prin
ceton University\n\nA rigorous relation is described between an interestin
g class of resonance motions in the gravitational three-body problem that
transition between different resonance orbits\, with Schrodinger’s wave eq
uation that models energy transitions of an electron. This is done via wea
k capture and escape motions. In this way\, problems in celestial mechanic
s can be studied using quantum mechanics methods.
DTSTART:20200326T163000Z
DTEND:20200326T173000Z
LAST-MODIFIED:20200323T142732Z
LOCATION:Remote Access
SUMMARY:Princeton University Thunch Talk - Added
URL:https://www.ias.edu/node/110196
END:VEVENT
BEGIN:VEVENT
UID:5bd29ee6-d900-4566-b856-4ab319dccce3
DTSTAMP:20200602T173219Z
CREATED:20200324T130757Z
DESCRIPTION:Topic: Wiles defect for Hecke algebras that are not complete in
tersections\n\nSpeaker: Chandrashekhar Khare\, University of California\n
\nIn his work on modularity theorems\, Wiles proved a numerical criterion
for a map of rings R->T to be an isomorphism of complete intersections. In
addition to proving modularity theorems\, this numerical criterion also i
mplies a connection between the order of a certain Selmer group and a spec
ial value of an L-function.\n\nIn this talk I will consider the case of a
Hecke algebra acting on the cohomology a Shimura curve associated to a qua
ternion algebra. In this case\, one has an analogous map of rings R->T whi
ch is known to be an isomorphism\, but in many cases the rings R and T fai
l to be complete intersections. This means that Wiles's numerical criterio
n will fail to hold.\n\nI will describe a method for precisely computing t
he extent to which the numerical criterion fails (i.e. the 'Wiles defect')
at a newform f which gives rise to an augmentation T -> Z_p. The defect t
urns out to be determined entirely by local information at the primes q di
viding the discriminant of the quaternion algebra at which the mod p repre
sentation arising from f is ``trivial''. (For instance if f corresponds to
a semistable elliptic curve\, then the local defect at q is related to th
e ``tame regulator'' of the Tate period of the elliptic curve at q.)\n\nTh
is is joint work with Gebhard Boeckle and Jeffrey Manning.
DTSTART:20200326T203000Z
DTEND:20200326T213000Z
LAST-MODIFIED:20200325T165543Z
LOCATION:https://theias.zoom.us/j/280491607
SUMMARY:Joint IAS/Princeton University Number Theory Seminar
URL:https://www.ias.edu/node/110231
END:VEVENT
BEGIN:VEVENT
UID:8f351734-eb15-4811-8192-3944e39cdb2a
DTSTAMP:20200602T173219Z
CREATED:20200327T115425Z
DESCRIPTION:Topic: Fragmentation pseudo-metrics and Lagrangian submanifolds
\n\nSpeaker: Octav Cornea \, Université de Montréal\n\nVideo: https://vide
o.ias.edu/Symplectic/2020/0327-OctavCornea\n\nThe purpose of the talk is t
o discuss a class of pseudo-metrics that can be defined on the set of obje
cts of a triangulated category whose morphisms are endowed with a notion o
f weight. In case the objects are Lagrangian submanifolds (possibly immers
ed) there are a some natural ways to define such pseudo-metrics and\, if t
he class of Lagrangian submanifolds is unobstructed\, these pseudo-metrics
are non-degenerate and extend in a natural way the Hofer distance.\n\nThe
talk is based on joint work with P. Biran and with E. Shelukhin.
DTSTART:20200327T130000Z
DTEND:20200327T140000Z
LAST-MODIFIED:20200401T002049Z
LOCATION:https://zoom.us/j/496042680
SUMMARY:Symplectic Seminar
URL:https://www.ias.edu/node/110276
END:VEVENT
BEGIN:VEVENT
UID:b4165507-a3b2-4b83-a661-8fbb5e139f30
DTSTAMP:20200602T173219Z
CREATED:20200128T185308Z
DESCRIPTION:Topic: General Discussion\n\n
DTSTART:20200327T190000Z
DTEND:20200327T200000Z
LAST-MODIFIED:20200327T131955Z
LOCATION:Virtual Meeting
SUMMARY:Princeton University HSC Group Discussion
URL:https://www.ias.edu/node/108286
END:VEVENT
BEGIN:VEVENT
UID:00505468-a6c9-4417-8559-479d71fb5ae2
DTSTAMP:20200602T173219Z
CREATED:20190607T202005Z
DESCRIPTION:Topic: CSPs with Global Modular Constraints: Algorithms and Har
dness via Polynomial Representations\n\nSpeaker: Sivakanth Gopi\n\nVideo:
https://video.ias.edu/csdm/2020/0330-SivakanthGopi\n\nA theorist's dream i
s to show that hard instances/obstructions for an (optimal) algorithm can
be used as gadgets to prove tight hardness reductions (which proves optima
lity of the algorithm). An example of such a result is that of Prasad Ragh
avendra who showed that for any constraint satisfaction problem (CSP)\, th
ere is an SDP which achieves the best possible approximation factor assumi
ng UGC. We show that a similar phenomenon occurs in CSPs with global modul
ar constraints.\n\nA global modular constraint is a linear equation (in al
l the variables) modulo M for some fixed constant M. Take any polytime sol
vable Boolean CSP like 2SAT or LIN_2 or HORNSAT. Let's look at LIN_2 for c
oncreteness\, it is a system of linear equations modulo 2. By Gaussian eli
mination over F_2\, we can find in polynomial time if it is satisfiable. N
ow suppose we are given an additional linear equation modulo M (for some f
ixed constant M not equal to 2). Can we find in polynomial time if the new
system is satisfiable? Surprisingly\, we show that the answer depends on
the prime factorization of M! For example\, it is possible for M=3\, but n
ot for M=15. We show that for such problems\, the obstructions to a natura
l algorithm and gadgets useful for hardness reduction (assuming ETH) are c
losely connected to complexity (like degree or sparsity) of polynomial rep
resentations of OR mod M. Thus showing good lower bounds on the complexity
of such polynomials imply good algorithms and constructing low complexity
representations imply good hardness results. We also show some connection
s to submodular minimization with global modular constraints.\n\nJoint wor
k with Joshua Brakensiek and Venkatesan Guruswami.
DTSTART:20200330T150000Z
DTEND:20200330T160000Z
LAST-MODIFIED:20200401T002203Z
LOCATION:https://theias.zoom.us/j/360043913
SUMMARY:Computer Science/Discrete Mathematics Seminar I
URL:https://www.ias.edu/node/100671
END:VEVENT
BEGIN:VEVENT
UID:e37b80d7-556b-40bf-8be6-6a9f603aceac
DTSTAMP:20200602T173219Z
CREATED:20190416T134501Z
DESCRIPTION:Topic: Blinded Challenge for Precision Cosmology with Galaxy Su
rveys\n\nSpeaker: Mikhail Ivanov & Marko Simonović\, New York University\;
CERN (respectively)\n\n
DTSTART:20200330T163000Z
DTEND:20200330T180000Z
LAST-MODIFIED:20200330T133057Z
LOCATION:Remote Access Meeting
SUMMARY:Institute for Advanced Study/Princeton University Early Universe/Co
smology Lunch Discussion
URL:https://www.ias.edu/node/99151
END:VEVENT
BEGIN:VEVENT
UID:185f56d6-5e97-4b4f-bfdd-106ea15061e3
DTSTAMP:20200602T173219Z
CREATED:20190607T203002Z
DESCRIPTION:Topic: High dimensional expansion and agreement testing\n\nSpea
ker: Irit Dinur\, Weizmann Institute of Science\; Visiting Professor\, Sch
ool of Mathematics\n\nVideo: https://video.ias.edu/csdm/2020/0331-IritDinu
r\n\nIn this talk I will describe the notion of 'agreement tests' that are
motivated by PCPs but stand alone as a combinatorial property-testing que
stion. I will show that high dimensional expanders support agreement tests
\, thereby derandomizing direct product tests in a very strong way.
DTSTART:20200331T143000Z
DTEND:20200331T163000Z
LAST-MODIFIED:20200401T002303Z
LOCATION:https://theias.zoom.us/j/360043913
SUMMARY:Computer Science/Discrete Mathematics Seminar II
URL:https://www.ias.edu/node/100686
END:VEVENT
BEGIN:VEVENT
UID:f88a209f-c853-453f-81a3-dd84f37d256e
DTSTAMP:20200602T173219Z
CREATED:20200414T122552Z
DESCRIPTION:Topic: Some Recent Insights on Transfer Learning\n\nSpeaker: Sa
mory Kpotufe\, Columbia University\; Member\, School of Mathematics\n\nA c
ommon situation in Machine Learning is one where training data is not full
y representative of a target population due to bias in the sampling mechan
ism or high costs in sampling the target population\; in such situations\,
we aim to ’transfer’ relevant information from the training data (a.k.a.
source data) to the target application. How much information is in the sou
rce data? How much target data should we collect if any? These are all pra
ctical questions that depend crucially on 'how far' the source domain is f
rom the target. However\, how to properly measure 'distance' between sourc
e and target domains remains largely unclear.\n\nIn this talk we will argu
e that much of the traditional notions of 'distance' (e.g. KL-divergence\,
extensions of TV such as D_A discrepancy\, density-ratios\, Wasserstein d
istance) can yield an over-pessimistic picture of transferability. Instead
\, we show that some new notions of 'relative dimension' between source an
d target (which we simply term 'transfer-exponents') capture a continuum f
rom easy to hard transfer. Transfer-exponents uncover a rich set of situat
ions where transfer is possible even at fast rates\, helps answer question
s such as the benefit of unlabeled or labeled target data\, yields a sense
of optimal vs suboptimal transfer heuristics\, and have interesting impli
cations for related problems such as multi-task learning.\n\nFinally\, tra
nsfer-exponents provide guidance as to *how* to efficiently sample target
data so as to guarantee improvement over source data alone. We illustrate
these new insights through various simulations on controlled data\, and on
the popular CIFAR-10 image dataset.\n\nThe talk is based on work with Gui
llaume Martinet\, and ongoing work with Steve Hanneke.
DTSTART:20200331T160000Z
DTEND:20200331T173000Z
LAST-MODIFIED:20200414T122552Z
LOCATION:https://theias.zoom.us/j/384099138
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/110511
END:VEVENT
BEGIN:VEVENT
UID:95f37c4b-965e-4a32-bbcf-ce9bd77bf2d1
DTSTAMP:20200602T173219Z
CREATED:20191025T210001Z
DESCRIPTION:Topic: Learning Controllable Representations\n\nSpeaker: Richar
d Zemel\, University of Toronto\; Member\, School of Mathematics\n\nVideo:
https://video.ias.edu/machinelearning/2020/0402-RichardZemel\n\nAs deep l
earning systems become more prevalent in real-world applications it is ess
ential to allow users to exert more control over the system. Exerting some
structure over the learned representations enables users to manipulate\,
interpret\, and even obfuscate the representations\, and may also improve
out-of-distribution generalization. In this talk I will discuss recent wor
k that makes some steps towards these goals\, aiming to represent the inpu
t in a factorized form\, with dimensions of the latent space partitioned i
nto task-dependent and task-independent components. I will focus on an app
roach that utilizes a reversible formulation of a network. Doing this reve
als a lack of robustness: the system is too invariant to a wide range of t
ask-relevant changes\, leading to a novel form of adversarial attacks whic
h we term excessive invariance. Our main contribution is an information-th
eoretic objective that encourages the model to develop factorized represen
tations. This provides the first approach tailored explicitly to overcome
excessive invariance and resulting vulnerabilities. It also facilitates se
veral applications\, including domain adaptation.
DTSTART:20200402T160000Z
DTEND:20200402T173000Z
LAST-MODIFIED:20200403T140032Z
LOCATION:https://theias.zoom.us/j/384099138
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/106011
END:VEVENT
BEGIN:VEVENT
UID:ba08ad62-d7dc-4b7b-b596-3bd26551889b
DTSTAMP:20200602T173219Z
CREATED:20200326T200016Z
DESCRIPTION:Topic: Secondary Acceleration in Magnetic Reconnection\n\nSpeak
er: Hayk Hakobyan\, Princeton University Graduate Student\n\nMagnetic reco
nnection has been invoked to explained high energy non-thermal radiation f
rom a myriad of astrophysical objects varying from blazar jets and black h
ole accretion disks\, to pulsar magnetospheres and solar flares. Although
the fundamentals of this process have been intensively studied over the pa
st two decades using various computational techniques\, there are still ma
ny open questions regarding the particle acceleration mechanisms. In this
talk I will briefly cover what we already know about the properties of the
relativistic magnetic reconnection. Followed by that I will present our r
ecent results in studying the previously undiscussed long-term (secondary)
acceleration process and the formation of the power-law distribution func
tion. This novel particle energization channel can be important in certain
astrophysical systems\, such as blazar jets\, where a broken power-law di
stribution of leptons is required to explain the observed radiation.
DTSTART:20200402T170000Z
DTEND:20200402T180000Z
LAST-MODIFIED:20200402T132025Z
LOCATION:Remote Access
SUMMARY:Princeton University Thunch Talk - Start Time 1 pm!
URL:https://www.ias.edu/node/110251
END:VEVENT
BEGIN:VEVENT
UID:c83c4b08-1c43-40a5-8c10-69ca24461c58
DTSTAMP:20200602T173219Z
CREATED:20190607T212006Z
DESCRIPTION:Topic: Density conjecture for horizontal families of lattices i
n SL(2)\n\nSpeaker: Mikolaj Fraczyk\, Member\, School of Mathematics\n\nVi
deo: https://video.ias.edu/puias/2020/0402-MikolajFraczyk\n\nLet G be a re
al semi-simple Lie group with an irreducible unitary representation \pi. T
he non-temperedness of \pi is measured by the parameter p(\pi) which is de
fined as the infimum of p\geq 2 such that \pi has matrix coefficients in L
^p(G). Sarnak and Xue conjectured that for any arithmetic lattice \Gamma \
subset G and principal congruence subgroup \Gamma(q)\subset \Gamma\, the m
ultiplicity of \pi in L^2(G/\Gamma(q)) is at most O(V(q)^{2/p(\pi)+\epsilo
n}) where V(q) is the covolume of \Gamma(q). In some contexts such estimat
e is a decent substitute for the Ramanujan conjecture. For G of real rank
1 Sarnak and Xue translate the estimate into a diophantine counting proble
m which they managed to solve SL(2\,R) and SL(2\,C).\n\nIn this talk I wil
l explain how one can get the same multiplicity bounds for families of pai
rwise non-commensurable lattices in G=SL(2\,R)\,SL(2\,C) given as unit gro
ups of maximal orders of quaternion algebras over number fields (“horizont
al families”). Namely: m(\pi\,\Gamma)\n\nTalk is based on a joint work wit
h Gergely Harcos\, Peter Maga and Djordje Milicevic.
DTSTART:20200402T203000Z
DTEND:20200402T213000Z
LAST-MODIFIED:20200403T140110Z
LOCATION:https://theias.zoom.us/j/959183254
SUMMARY:Joint IAS/Princeton University Number Theory Seminar
URL:https://www.ias.edu/node/101361
END:VEVENT
BEGIN:VEVENT
UID:a3254257-05ce-46ce-800d-bcae4e2299db
DTSTAMP:20200602T173219Z
CREATED:20200415T142632Z
DESCRIPTION:Topic: The Simplicity Conjecture\n\nSpeaker: Daniel Cristofaro-
Gardiner\, Member\, School of Mathematics\n\nI will explain recent joint w
ork proving that the group of compactly supported area preserving homeomor
phisms of the two-disc is not a simple group\; this answers the ”Simplicit
y Conjecture” in the affirmative. Our proof uses new spectral invariants\,
defined via periodic Floer homology\, that I will introduce: these recove
r the Calabi invariant of monotone twists.
DTSTART:20200403T130000Z
DTEND:20200403T140000Z
LAST-MODIFIED:20200415T144528Z
LOCATION:https://zoom.us/j/496042680
SUMMARY:Symplectic Seminar
URL:https://www.ias.edu/node/110546
END:VEVENT
BEGIN:VEVENT
UID:2b83a8a8-aad8-4da4-ad58-6a2c2f60001b
DTSTAMP:20200602T173219Z
CREATED:20200128T185339Z
DESCRIPTION:Topic: General Discussion\n\n
DTSTART:20200403T190000Z
DTEND:20200403T200000Z
LAST-MODIFIED:20200402T141037Z
LOCATION:Remote Event
SUMMARY:Princeton University HSC Group Discussion
URL:https://www.ias.edu/node/108291
END:VEVENT
BEGIN:VEVENT
UID:f9f29233-bce0-4ff6-9355-be68ae81dca1
DTSTAMP:20200602T173219Z
CREATED:20200403T114109Z
DESCRIPTION:Topic: Borrowing memory that's being used: catalytic approaches
to the Tree Evaluation Problem\n\nSpeaker: James Cook\, University of Tor
onto\n\nVideo: https://video.ias.edu/csdm/2020/0406-JamesCook\n\nI'll be p
resenting some joint work with Ian Mertz scheduled to appear at STOC 2020.
\n\nThe study of the Tree Evaluation Problem (TEP)\, introduced by S. Cook
et al. (TOCT 2012)\, is a promising approach to separating L from P. Give
n a label in [k] at each leaf of a complete binary tree and an explicit fu
nction [k]^2 -> [k] for recursively computing the value of each internal n
ode from its children\, the problem is to compute the value at the root no
de. A simple recursive algorithm can solve this using Theta(h log k) memor
y\, where h is the tree height and k is the size of the alphabet. Until no
w\, no better deterministic algorithm was known.\n\nWe present a new algor
ithm which uses less memory when k is not too big compared to h\, inspired
by a surprising result from Burhman et al. about “catalytic space' comput
ation (STOC 2012). Ours is the first algorithm to beat the simple recursiv
e algorithm\, and also the first non-trivial approach to proving a determi
nistic upper bound for TEP.
DTSTART:20200406T150000Z
DTEND:20200406T160000Z
LAST-MODIFIED:20200407T135221Z
LOCATION:https://theias.zoom.us/j/360043913
SUMMARY:Computer Science/Discrete Mathematics Seminar I
URL:https://www.ias.edu/node/110381
END:VEVENT
BEGIN:VEVENT
UID:31180f04-9678-49cd-9b06-cd7fa6922c69
DTSTAMP:20200602T173219Z
CREATED:20190802T151502Z
DESCRIPTION:Topic: Perfectly Parallel Cosmological Simulations Using Spatia
l Comoving Lagrangian Acceleration\n\nSpeaker: Florent Leclercq\, Imperial
College\n\nI will introduce a new\, perfectly parallel approach to simula
te cosmic structure formation\, based on the spatial COmoving Lagrangian A
cceleration (sCOLA) framework. Building upon a hybrid analytical and numer
ical description of particles' trajectories\, sCOLA allows an efficient ti
ling of a cosmological volume\, where the dynamics within each tile is com
puted independently. I will show that cosmological simulations at the degr
ee of accuracy required for the analysis of the next generation of surveys
can be run in drastically reduced wall-clock times and with very low memo
ry requirements\, and discuss perspectives for computing future larger and
higher-resolution cosmological simulations\, taking advantage of a variet
y of hardware architectures.
DTSTART:20200406T163000Z
DTEND:20200406T180000Z
LAST-MODIFIED:20200402T133914Z
LOCATION:Until further notice\, meetings will be held remotely on Zoom.
SUMMARY:Institute for Advanced Study/Princeton University Early Universe/Co
smology Lunch Discussion
URL:https://www.ias.edu/node/102636
END:VEVENT
BEGIN:VEVENT
UID:9b1dc40b-f91e-42ad-a7af-cb66d1f163ab
DTSTAMP:20200602T173219Z
CREATED:20200327T183132Z
DESCRIPTION:Topic: The Palais-Smale Theorem and the Solution of Hilbert’s 2
3 Problem\n\nSpeaker: Karen Uhlenbeck\, The University of Texas at Austin\
; Distinguished Visiting Professor\, School of Mathematics\n\nVideo: https
://video.ias.edu/members/2020/0406-KarenUhlenbeck\n\nHilbert’s 23rd Proble
m is the last in his famous list of problems and is of a different charact
er than the others. The description is several pages\, and basically says
that the calculus of variations is a subject which needs development. We w
ill look in retrospect at one of the critical events in the calculus of va
riations: The point at which the critical role of dimension was understood
\, and the role that the Palais-Smale condition(1963) played in this under
standing. I apologize that in its present state\, the talk consists mostly
of my reminiscences and lacks references. I welcome suggestions from the
audience.
DTSTART:20200406T180000Z
DTEND:20200406T190000Z
LAST-MODIFIED:20200407T135313Z
LOCATION:http://theias.zoom.us/j/119412864
SUMMARY:Members' Seminar
URL:https://www.ias.edu/node/110281
END:VEVENT
BEGIN:VEVENT
UID:f9c9cd92-f74e-46e5-9f7b-af74d15d49e6
DTSTAMP:20200602T173219Z
CREATED:20190607T203002Z
DESCRIPTION:Topic: Primality testing\n\nSpeaker: Andrey Kupavskii \, Member
\, School of Mathematics\n\nVideo: https://video.ias.edu/csdm/2020/0407-An
dreyKupavskii\n\nIn the talk\, I will explain the algorithm (and its analy
sis) for testing whether a number is a prime\, invented by Agrawal\, Kayal
\, and Saxena.
DTSTART:20200407T143000Z
DTEND:20200407T163000Z
LAST-MODIFIED:20200408T163016Z
LOCATION:https://theias.zoom.us/j/360043913
SUMMARY:Computer Science/Discrete Mathematics Seminar II
URL:https://www.ias.edu/node/100681
END:VEVENT
BEGIN:VEVENT
UID:a956b422-3bff-4f8f-b8f9-0e813ab6854b
DTSTAMP:20200602T173219Z
CREATED:20200326T132006Z
DESCRIPTION:Topic: Interpolation in learning: steps towards understanding w
hen overparameterization is harmless\, when it helps\, and when it causes
harm\n\nSpeaker: Anant Sahai\, University of California\, Berkeley\n\nVide
o: https://video.ias.edu/machinelearning/2020/0407-AnantSahai\n\nA continu
ing mystery in understanding the empirical success of deep neural networks
has been in their ability to achieve zero training error and yet generali
ze well\, even when the training data is noisy and there are many more par
ameters than data points. Following the information-theoretic tradition of
seeking understanding\, this talk will share our four-part approach to sh
edding some light on this phenomenon. First\, following the tradition of s
uch distilled toy models like the BSC and AWGN channels\, the Gaussian sou
rce\, or scalar linear control systems\, we zoom in on the classical linea
r regression problem in the underdetermined setting with more parameters t
han training data. Here\, the solutions that minimize training error inter
polate the data\, including noise. Second\, following the tradition of con
verse bounds\, we give a genie-aided bound on how well any interpolative s
olutions can generalize to fresh test data\, and show that this bound gene
rically decays to zero (at a known rate) with the number of extra features
\, thus characterizing an explicit benefit of overparameterization. Third\
, we talk about what it takes to achieve such harmless interpolation in ap
propriately overparameterized limits. The key for us turns out to be some
appropriate sense of implicit or explicit sparsity. Along the way\, we cal
l out a family of key concepts that help us understand what is required to
achieve harmless interpolation in general\, in particular the ideas of 's
ignal survival' and 'contamination.' Fourth\, we take what we have learned
from the regression setting to better understand overparameterized classi
fication\, and see how there is a regime in which classification is fundam
entally more forgiving than regression\, though it can be understood using
the same tools.
DTSTART:20200407T160000Z
DTEND:20200407T173000Z
LAST-MODIFIED:20200408T163041Z
LOCATION:https://theias.zoom.us/j/384099138
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/110246
END:VEVENT
BEGIN:VEVENT
UID:2172c9f0-7817-4beb-b2d2-95bc857742a7
DTSTAMP:20200602T173219Z
CREATED:20190514T144502Z
DESCRIPTION:Institute Film Series screening of a film curated by the School
of Social Science.
DTSTART:20200408T200000Z
DTEND:20200408T223000Z
LAST-MODIFIED:20200316T162035Z
LOCATION:Wolfensohn Hall
SUMMARY:Canceled: Institute Film Series
URL:https://www.ias.edu/node/100061
END:VEVENT
BEGIN:VEVENT
UID:6c5d82e3-20a7-405e-a595-8cdebd7f07a5
DTSTAMP:20200602T173219Z
CREATED:20200403T115532Z
DESCRIPTION:Topic: Meta-Learning: Why It’s Hard and What We Can Do\n\nSpeak
er: Ke Li\, Member\, School of Mathematics\n\nVideo: https://video.ias.edu
/machinelearning/2020/0409-KeLi\n\nMeta-learning (or learning to learn) st
udies how to use machine learning to design machine learning methods thems
elves. We consider an optimization-based formulation of meta-learning that
learns to design an optimization algorithm automatically\, which we call
Learning to Optimize. Surprisingly\, it turns out that the most straightfo
rward approach of learning such an algorithm\, namely backpropagation\, do
es not work. We explore the underlying reason for this failure\, devise a
solution based on reinforcement learning and discuss the open challenges i
n meta-learning.
DTSTART:20200409T190000Z
DTEND:20200409T203000Z
LAST-MODIFIED:20200410T142227Z
LOCATION:https://theias.zoom.us/j/384099138
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/110386
END:VEVENT
BEGIN:VEVENT
UID:0f5398e8-5e94-409c-9245-0347c4f0dfc3
DTSTAMP:20200602T173219Z
CREATED:20200403T120214Z
DESCRIPTION:Topic: On the Kudla-Rapoport conjecture\n\nSpeaker: Chao Li\, C
olumbia University\n\nVideo: https://video.ias.edu/puias/2020/0409-ChaoLi
\n\nThe Kudla-Rapoport conjecture predicts a precise identity between the
arithmetic intersection number of special cycles on unitary Rapoport-Zink
spaces and the derivative of local representation densities of hermitian f
orms. It is a key local ingredient to establish the arithmetic Siegel-Weil
formula and the arithmetic Rallis inner product formula\, relating the he
ight of special cycles on Shimura varieties to the derivative of Siegel Ei
senstein series and L-functions. We will motivate this conjecture\, explai
n a proof and discuss global applications. This is joint work with Wei Zha
ng.
DTSTART:20200409T203000Z
DTEND:20200409T213000Z
LAST-MODIFIED:20200410T142257Z
LOCATION:https://theias.zoom.us/j/959183254
SUMMARY:Joint IAS/Princeton University Number Theory Seminar
URL:https://www.ias.edu/node/110391
END:VEVENT
BEGIN:VEVENT
UID:9319df60-6f4f-490e-ac8f-cac90a63bb54
DTSTAMP:20200602T173219Z
CREATED:20200414T144248Z
DESCRIPTION:Topic: Consequences of delays and imperfect isolation in epidem
ic control\n\nSpeaker: Lai-Sang Young \, New York University\; Member\, Sc
hool of Mathematics\n\nIn the absence of a vaccine\, isolation is about th
e only available means to control an epidemic. I would like to share with
everyone some things I learned from a project I worked on a few years ago
studying the consequences of delays and imperfect isolation. We used a toy
model defined by a system of delay differential equations (which gives ri
se to an infinite dimensional dynamical system) to illustrate some simple
points. Some of the ideas may resonate\, but I stress that this is NOT spe
cifically about covid-19.
DTSTART:20200409T213000Z
DTEND:20200409T230000Z
LAST-MODIFIED:20200414T144619Z
LOCATION:Remote Access Only
SUMMARY:Mathematical Conversations
URL:https://www.ias.edu/node/110521
END:VEVENT
BEGIN:VEVENT
UID:3cc9f6f6-4d1f-45c6-9f30-1186c10a71b7
DTSTAMP:20200602T173219Z
CREATED:20200410T145741Z
DESCRIPTION:Topic: Geometry of Quantum Uncertainty\n\nSpeaker: Leonid Polte
rovich\, Tel Aviv University\n\nVideo: https://video.ias.edu/symplectic/20
20/0410-LeonidPolterovich\n\nCompatible almost-complex structures on sympl
ectic manifolds correspond to optimal quantizations. I will discuss this s
tatement (joint with Louis Ioos and David Kazhdan)\, as well as some other
geometric facets of uncertainty principles in quantum mechanics.
DTSTART:20200410T130000Z
DTEND:20200410T140000Z
LAST-MODIFIED:20200410T201602Z
LOCATION:https://princeton.zoom.us/j/745635914
SUMMARY:Symplectic Seminar
URL:https://www.ias.edu/node/110446
END:VEVENT
BEGIN:VEVENT
UID:22e85309-304e-410b-aaf5-57b25607ba25
DTSTAMP:20200602T173219Z
CREATED:20200402T141112Z
DESCRIPTION:Topic: General Discussion\n\n
DTSTART:20200410T190000Z
DTEND:20200410T200000Z
LAST-MODIFIED:20200402T141112Z
LOCATION:Remote Event
SUMMARY:Princeton University HSC Group Discussion
URL:https://www.ias.edu/node/110326
END:VEVENT
BEGIN:VEVENT
UID:34207212-c835-4976-8344-da1b213eeaf2
DTSTAMP:20200602T173219Z
CREATED:20200408T154940Z
DESCRIPTION:Topic: Flows of vector fields: classical and modern\n\nSpeaker:
Camillo DeLellis\, IBM von Neumann Professor\, School of Mathematics\n\nV
ideo: https://video.ias.edu/analysis/2020/0413-CamilloDeLellis\n\nConsider
a (possibly time-dependent) vector field $v$ on the Euclidean space. The
classical Cauchy-Lipschitz (also named Picard-Lindel\'of) Theorem states t
hat\, if the vector field $v$ is Lipschitz in space\, for every initial da
tum $x$ there is a unique trajectory $\gamma$ starting at $x$ at time $0$
and solving the ODE $\dot{\gamma} (t) = v (t\, \gamma (t))$. The theorem l
ooses its validity as soon as $v$ is slightly less regular. However\, if w
e bundle all trajectories into a global map allowing $x$ to vary\, a celeb
rated theory put forward by DiPerna and Lions in the 80es show that there
is a unique such flow under very reasonable conditions and for much less r
egular vector fields. A long-standing open question is whether this theory
is the byproduct of a stronger classical result which ensures the uniquen
ess of trajectories for {\em almost every} initial datum. I will give a co
mplete answer to the latter question and draw connections with partial dif
ferential equations\, harmonic analysis\, probability theory and Gromov's
$h$-principle.
DTSTART:20200413T150000Z
DTEND:20200413T160000Z
LAST-MODIFIED:20200413T194448Z
LOCATION:https://theias.zoom.us/j/373002666
SUMMARY:Analysis Seminar
URL:https://www.ias.edu/node/110421
END:VEVENT
BEGIN:VEVENT
UID:911a3bf4-5f0b-41af-b662-d9213bab962f
DTSTAMP:20200602T173219Z
CREATED:20200409T143458Z
DESCRIPTION:Topic: Legal Theorems of Privacy\n\nSpeaker: Kobbi Nissim\, Geo
rgetown University\n\nVideo: https://video.ias.edu/csdm/2020/0410-KobbiNis
sim\n\nThere are significant gaps between legal and technical thinking aro
und data privacy. Technical standards such as k-anonymity and differential
privacy are described using mathematical language whereas legal standards
are not rigorous from a mathematical point of view and often resort to co
ncepts such as de-identification and anonymization which they only partial
ly define. As a result\, arguments about the adequacy of technical privacy
measures for satisfying legal privacy often lack rigor\, and their conclu
sions are uncertain. The uncertainty is exacerbated by a litany of success
ful privacy attacks on privacy measures thought to meet legal expectations
but then shown to fall short of doing so. In this work\, we ask whether i
t is possible to introduce mathematical rigor into such analyses to the po
int of making and proving formal “legal theorems” that certain technical p
rivacy measures meet legal expectations. For that\, we explore some of the
gaps between these two very different approaches\, and present initial st
rategies towards bridging these gaps considering examples from US and EU l
aw.\n\nBased on work with Aloni Cohen
DTSTART:20200413T150000Z
DTEND:20200413T160000Z
LAST-MODIFIED:20200413T194553Z
LOCATION:https://theias.zoom.us/j/360043913
SUMMARY:Computer Science/Discrete Mathematics Seminar I
URL:https://www.ias.edu/node/110436
END:VEVENT
BEGIN:VEVENT
UID:3d6dfd54-7011-4fcc-b9e5-71c995e9685e
DTSTAMP:20200602T173219Z
CREATED:20190416T134501Z
DESCRIPTION:Topic: Description of Galaxy Distribution on Large Scales Utili
zing the Equivalence Principle\n\nSpeaker: Zvonimir Vlah\, CERN\n\nOn larg
e cosmological scales\, galaxies and other tracers of large scale structur
e can be organized in the perturbative bias expansion. After shortly revie
wing the canonical approaches relying on Eulerian and Lagrangian dynamics\
, I will describe a novel bias expansion scheme orthogonal to these existi
ng strategies. The new approach uses the consequences of the equivalence p
rinciple in n-point correlators in order to construct the physical galaxy
field on large scales. I will compare the obtained results to the current
literature and discuss the generalized ΛCDM background setting.
DTSTART:20200413T163000Z
DTEND:20200413T180000Z
LAST-MODIFIED:20200410T140916Z
LOCATION:Until further notice\, meetings will be held remotely on Zoom.
SUMMARY:Institute for Advanced Study/Princeton University Early Universe/Co
smology Lunch Discussion
URL:https://www.ias.edu/node/99156
END:VEVENT
BEGIN:VEVENT
UID:f70ddb6c-d895-4be4-970a-26527d834c01
DTSTAMP:20200602T173219Z
CREATED:20200408T131449Z
DESCRIPTION:Topic: Towards Robust Artificial Intelligence \n\nSpeaker: Push
meet Kohli\, Google\n\nVideo: https://video.ias.edu/workshop/2020/0415-16
\n\nDeep learning has led to rapid progress being made in the field of mac
hine learning and artificial intelligence\, leading to dramatically improv
ed solutions of many challenging problems such as image understanding\, sp
eech recognition\, and control systems. Despite these remarkable successes
\, researchers have observed some intriguing and troubling aspects of the
behaviour of these models. A case in point is the presence of adversarial
examples which make learning based systems fail in unexpected ways. Such b
ehaviour and the difficulty of interpreting the behaviour of neural networ
ks is a serious hindrance in the deployment of these models for safety-cri
tical applications. In this talk\, I will review the challenges in develop
ing models that are robust and explainable and discuss the opportunities f
or collaboration between the formal methods and machine learning communiti
es.
DTSTART:20200415T130000Z
DTEND:20200415T140000Z
LAST-MODIFIED:20200421T131219Z
LOCATION:Virtual
SUMMARY:Workshop on New Directions in Optimization\, Statistics and Machine
Learning
URL:https://www.ias.edu/node/110411
END:VEVENT
BEGIN:VEVENT
UID:ff604817-2335-4dc7-9d60-2e145eccbdee
DTSTAMP:20200602T173219Z
CREATED:20200408T132944Z
DESCRIPTION:Topic: A snapshot of few-shot classification\n\nSpeaker: Richar
d Zemel\, University of Toronto\; Visitor\, School of Mathematics\n\nVideo
: https://video.ias.edu/workshop/2020/0415-16\n\nFew-shot classification\,
the task of adapting a classifier to unseen classes given a small labeled
dataset\, is an important step on the path toward human-like machine lear
ning. I will present some of the key advances in this area\, and will then
focus on the fundamental issue of overfitting in the few-shot scenario. B
ayesian methods are well-suited to tackling this issue because they allow
practitioners to specify prior beliefs and update those beliefs in light o
f observed data. Contemporary approaches to Bayesian few-shot classificati
on maintain a posterior distribution over model parameters\, which is slow
and requires storage that scales with model size. Instead\, we propose a
Gaussian process classifier based on a novel combination of Pólya-gamma au
gmentation and the one-vs-each loss that allows us to efficiently marginal
ize over functions rather than model parameters. We demonstrate improved a
ccuracy and uncertainty quantification on both standard few-shot classific
ation benchmarks and few-shot domain transfer tasks. I will conclude by di
scussing open questions in this area.
DTSTART:20200415T141500Z
DTEND:20200415T151500Z
LAST-MODIFIED:20200421T131251Z
LOCATION:Virtual
SUMMARY:Workshop on New Directions in Optimization\, Statistics and Machine
Learning
URL:https://www.ias.edu/node/110416
END:VEVENT
BEGIN:VEVENT
UID:d094bfb1-1ccd-444d-8b89-602d787a9882
DTSTAMP:20200602T173219Z
CREATED:20200413T183205Z
DESCRIPTION:Topic: Studying Magnetic Fields and Dynamics of Star formation
with Velocity Gradients\n\nSpeaker: Alex Lazarian\, University of Wisconsi
n - Madison\n\nDue to the advances in understanding of MHD turbulence and
turbulent reconnection we formulated a new Velocity Gradient Technique (VG
T) of magnetic field tracing. The technique has been successfully tested u
sing high resolution numerical simulations as well as by comparing with ma
gnetic field maps obtained with the with Planck satellite and BLASTPOL pol
arimetry. In my talk I am going to focus on new prospects that the VGT pre
sents in terms of exploring (a) the 3D structure of magnetic fields in mol
ecular clouds\, (b) the magnetic connection of dense and diffuse gas and (
c) the dynamics of the gravitational collapse. The latter possibility is i
llustrated\, on the basis of numerical simulations\, in Figure 1. There it
is shown that how to identify using the VGT the regions of gravitational
collapse as well as to trace magnetic field. I will show of survey of near
by molecular clouds where both magnetic fields and regions of gravitationa
l collapse are identified using the VGT. In addition\, I shall show the 3D
magnetic structure of Vela C molecular cloud that is obtained by applying
the VGT to different molecular species that are formed at different optic
al depths. Finally\, I shall describe a new technique that demonstrates hi
gh precision in obtaining the strength of magnetic fields using the VGT. U
sing numerical data\, I will demonstrate that the new technique can provid
e the magnetic field strength in the regions strongly affected by self-gra
vity where the classical DavisChandrasekhar-Fermi technique fails. I will
show the results for magnetic field strength obtained for nearby molecular
clouds.
DTSTART:20200415T150000Z
DTEND:20200415T160000Z
LAST-MODIFIED:20200413T194949Z
LOCATION:Remote via Zoom
SUMMARY:Princeton University Star Formation/ISM Rendezvous (SFIR) - ADDED
URL:https://www.ias.edu/node/110506
END:VEVENT
BEGIN:VEVENT
UID:104d3dd8-13d1-4bdf-bd23-af0314bf5ddd
DTSTAMP:20200602T173219Z
CREATED:20200303T184406Z
DESCRIPTION:Topic: Iterative Random Forests (iRF) with applications to geno
mics and precision medicine \n\nSpeaker: Bin Yu\, University of Californi
a\, Berkeley\n\nVideo: https://video.ias.edu/workshop/2020/0415-16\n\nGeno
mics has revolutionized biology\, enabling the interrogation of whole tran
scriptomes\, genome-wide binding sites for proteins\, and many other molec
ular processes. However\, individual genomic assays measure elements that
interact in vivo as components of larger molecular machines. Understanding
how these high-order interactions drive gene expression presents a substa
ntial statistical challenge. Building on random forests (RFs) and random i
ntersection trees (RITs) and through extensive\, biologically inspired sim
ulations\, we developed the iterative random forest algorithm (iRF) to see
k predictable and stable high-order Boolean interactions. We demonstrate t
he utility of iRF for high-order Boolean interaction discovery in two pred
iction problems: enhancer activity in the early Drosophila embryo and red
hair phenotype using UK BioBank data. The latter is a proof-of-concept ste
p towards suggesting gene variants\nbehind cardiovasuclar phenotypes for s
ingle cell experiments as part of a Chan-Zuckerberg Biohub Intercampus Awa
rd to UC Berkeley\, UCSF and Stanford. Finally\, a connection is made betw
een iRF and our PCS framework where PCS stands for predictability\, comput
ability and Stability.
DTSTART:20200415T153000Z
DTEND:20200415T163000Z
LAST-MODIFIED:20200421T131332Z
LOCATION:Virtual
SUMMARY:Workshop on New Directions in Optimization\, Statistics and Machine
Learning
URL:https://www.ias.edu/node/109891
END:VEVENT
BEGIN:VEVENT
UID:a9927791-90fe-4d50-a9ec-429872d3eecf
DTSTAMP:20200602T173219Z
CREATED:20200303T183841Z
DESCRIPTION:Topic: Generative Modeling by Estimating Gradients of the Data
Distribution\n\nSpeaker: Stefano Ermon\, Stanford University\n\nVideo: htt
ps://video.ias.edu/workshop/2020/0415-16\n\nExisting generative models are
typically based on explicit representations of probability distributions
(e.g.\, autoregressive or VAEs) or implicit sampling procedures (e.g.\, GA
Ns). We propose an alternative approach based on modeling directly the vec
tor field of gradients of the data distribution (scores). Our framework al
lows flexible energy-based model architectures\, requires no sampling duri
ng training or the use of adversarial training methods. Using annealed Lan
gevin dynamics\, we produces samples comparable to GANs on MNIST\, CelebA
and CIFAR-10 datasets\, achieving a new state-of-the-art inception score o
f 8.91 on CIFAR-10. Finally\, I will discuss challenges in evaluating bias
and generalization in generative models.
DTSTART:20200415T180000Z
DTEND:20200415T190000Z
LAST-MODIFIED:20200421T131353Z
LOCATION:Virtual
SUMMARY:Workshop on New Directions in Optimization\, Statistics and Machine
Learning
URL:https://www.ias.edu/node/109876
END:VEVENT
BEGIN:VEVENT
UID:c70073bc-e952-49f8-a4db-23ea4a4f52cd
DTSTAMP:20200602T173219Z
CREATED:20200303T184115Z
DESCRIPTION:Topic: Interpretability for everyone\n\nSpeaker: Been Kim\, Goo
gle Brain\n\nVideo: https://video.ias.edu/workshop/2020/0415-16\n\nIn this
talk\, I would like to share some of my reflections on the progress made
in the field of interpretable machine learning. We will reflect on where w
e are going as a field\, and what are the things that we need to be aware
of to make progress. With that perspective\, I will then discuss some of m
y work on 1) sanity checking popular methods and 2) developing more lay pe
rson-friendly interpretability methods. I will also share some open theore
tical questions that may help us move forward. I hope this talk will offer
a new angle to look at ways to make progress in this field.\n\nA bit more
on 2): most of the recent interpretability methods are designed by machin
e learning experts\, for machine learning experts. l will introduce a diff
erent family of methods that are designed to help lay people - people who
may have non-ML domain expertise (e.g.\, medical). Testing with concept ac
tivation vectors (TCAV) is a post-training interpretability method for com
plex models\, such as neural networks. This method provides an interpretat
ion of a neural net's internal state in terms of human-friendly\, high-lev
el concepts instead of low-level input features. In other words\, this met
hod can “speak” the user’s language\, rather than the computer’s language.
I will also discuss more recent extensions of this work towards automatic
ally discovering high-level concepts as well as discovering ‘complete’ and
causal concepts.
DTSTART:20200415T191500Z
DTEND:20200415T201500Z
LAST-MODIFIED:20200421T131412Z
LOCATION:Virtual
SUMMARY:Workshop on New Directions in Optimization\, Statistics and Machine
Learning
URL:https://www.ias.edu/node/109881
END:VEVENT
BEGIN:VEVENT
UID:eee7e654-27dd-4e06-bbe5-506041a8ba07
DTSTAMP:20200602T173219Z
CREATED:20200303T184951Z
DESCRIPTION:Topic: Evaluating Lossy Compression Rates of Deep Generative Mo
dels\n\nSpeaker: Roger Grosse\, Member\, School of Mathematics\n\nVideo: h
ttps://video.ias.edu/workshop/2020/0415-16\n\nImplicit generative models s
uch as GANs have achieved remarkable progress at generating convincing fak
e images\, but how well do they really match the distribution? Log-likelih
ood has been used extensively to evaluate generative models whenever it’s
convenient to do so\, but measuring log-likelihoods for implicit generativ
e models presents computational challenges. Furthermore\, in order to obta
in a density\, one needs to smooth the distribution using a noisy model (t
ypically Gaussian)\, and this choice is hard to motivate. We take a differ
ent approach: viewing log-likelihood as a measure of lossless compression\
, we instead evaluate the lossy compression rates of the generative model\
, thereby removing the need for a noise distribution.\n\nWe show how a sin
gle run of annealed importance sampling (AIS) can be used to upper bound t
he entire rate-distortion curve for an implicit generative model. Interest
ingly\, with a Euclidean distortion metric\, this is a nearly identical co
mputation to how one would estimate a single scalar log-likelihood. But th
e rate-distortion curve gives a more complete picture of the performance o
f the model. We estimate rate-distortion curves for VAEs\, GANs\, and adve
rsarial autoencoders\, and arrive at insights not obtainable from log-like
lihoods alone.
DTSTART:20200415T203000Z
DTEND:20200415T213000Z
LAST-MODIFIED:20200421T131431Z
LOCATION:Virtual
SUMMARY:Workshop on New Directions in Optimization\, Statistics and Machine
Learning
URL:https://www.ias.edu/node/109916
END:VEVENT
BEGIN:VEVENT
UID:321d5d63-3abb-474f-a0c7-fe6099fa9921
DTSTAMP:20200602T173219Z
CREATED:20200424T180537Z
DESCRIPTION:Topic: Vignettes about pure mathematics and machine learning\n
\nSpeaker: Jordan Ellenberg\, University of Wisconsin-Madison\n\nThrough i
nteractions with engineers and computer scientists over the years\, includ
ing some current visitors at IAS\, I have become pretty sold on the idea t
hat machine learning is rich in questions which are interesting to pure ma
thematicians and where our insights might perhaps be useful. I will tell a
few short stories along these lines. The talk will not require expertise
in machine learning from either the audience or the speaker.
DTSTART:20200415T213000Z
DTEND:20200415T230000Z
LAST-MODIFIED:20200424T180537Z
LOCATION:Remote Access Only
SUMMARY:Mathematical Conversations
URL:https://www.ias.edu/node/110786
END:VEVENT
BEGIN:VEVENT
UID:eb584b1e-1f8e-4c2d-91ce-dc0070b4828c
DTSTAMP:20200602T173219Z
CREATED:20200303T185742Z
DESCRIPTION:Topic: Do Simpler Models Exist and How Can We Find Them?\n\nSpe
aker: Cynthia Rudin\, Duke University\n\nVideo: https://video.ias.edu/work
shop/2020/0415-16\n\nWhile the trend in machine learning has tended toward
s more complex hypothesis spaces\, it is not clear that this extra complex
ity is always necessary or helpful for many domains. In particular\, model
s and their predictions are often made easier to understand by adding inte
rpretability constraints. These constraints shrink the hypothesis space\;
that is\, they make the model simpler. Statistical learning theory suggest
s that generalization may be improved as a result as well. However\, addin
g extra constraints can make optimization (exponentially) harder. For inst
ance it is much easier in practice to create an accurate neural network th
an an accurate and sparse decision tree. We address the following question
: Can we show that a simple-but-accurate machine learning model might exis
t for our problem\, before actually finding it? If the answer is promising
\, it would then be worthwhile to solve the harder constrained optimizatio
n problem to find such a model. In this talk\, I present an easy calculati
on to check for the possibility of a simpler model. This calculation indic
ates that simpler-but-accurate models do exist in practice more often than
you might think. Time-permitting\, I will then briefly overview our progr
ess towards the challenging problem of finding optimal sparse decision tre
es.\n\nThis is joint work with Lesia Semenova\, Ron Parr\, Xiyang Hu\, Chu
di Zhong\, Jimmy Lin\, and Margo Seltzer.
DTSTART:20200416T140000Z
DTEND:20200416T150000Z
LAST-MODIFIED:20200421T131450Z
LOCATION:Virtual
SUMMARY:Workshop on New Directions in Optimization\, Statistics and Machine
Learning
URL:https://www.ias.edu/node/109956
END:VEVENT
BEGIN:VEVENT
UID:7f55e9d4-cc78-4cde-b8a7-b99d5cd4a61f
DTSTAMP:20200602T173219Z
CREATED:20200303T185821Z
DESCRIPTION:Topic: Deep equilibrium models via monotone operators\n\nSpeake
r: Zico Kolter\, Carnegie Mellon University\n\nVideo: https://video.ias.ed
u/workshop/2020/0415-16\n\nIn this talk\, I will first introduce our recen
t work on the Deep Equilibrium Model (DEQ). Instead of stacking nonlinear
layers\, as is common in deep learning\, this approach finds the equilibri
um point of the repeated iteration of a single non-linear layer\, then bac
kpropagates through the layer directly using the implicit function theorem
. The resulting method achieves or matches state of the art performance in
many domains (while consuming much less memory)\, and can theoretically e
xpress any 'traditional' deep network with just a single layer. However\,
existing work in DEQs leave open the question of the existence and uniquen
ess of these fixed points and attempts to compute them largely through heu
ristic methods. In the second part of the talk\, I will thus introduce a n
ew class of DEQ models based upon monotone operator theory. I will illustr
ate how we can parameterize these networks such that they are guaranteed t
o have a unique equilibrium point\, and show how to apply operator splitti
ng methods to efficiently find these fixed points. Finally\, I will close
by highlighting some open challenges in these areas.
DTSTART:20200416T151500Z
DTEND:20200416T161500Z
LAST-MODIFIED:20200421T131507Z
LOCATION:Virtual
SUMMARY:Workshop on New Directions in Optimization\, Statistics and Machine
Learning
URL:https://www.ias.edu/node/109961
END:VEVENT
BEGIN:VEVENT
UID:8eaaf1a0-0721-478f-9324-0a5b5ad56fd3
DTSTAMP:20200602T173219Z
CREATED:20200303T190224Z
DESCRIPTION:Topic: The Peculiar Optimization and Regularization Challenges
in Multi-Task Learning and Meta-Learning\n\nSpeaker: Chelsea Finn\, Stanfo
rd University and Google AI\n\nVideo: https://video.ias.edu/workshop/2020/
0415-16\n\nDespite the success of deep learning\, much of its success has
existed in settings where the goal is to learn one\, single-purpose functi
on from data. However\, in many contexts\, we hope to optimize neural netw
orks for multiple\, distinct tasks (i.e. multi-task learning)\, and optimi
ze so that what is learned from these tasks is transferable to the acquisi
tion of new tasks (e.g. as in meta-learning). In this talk\, I will discus
s how the multi-task and meta optimization problems differ from standard p
roblems\, and some of the unique challenges that arise\, both in optimizat
ion and in regularization. This includes (1) a kind of overfitting that is
unique to meta-learning\, where the optimizer memorizes not the labels\,
but the functions that solve the training tasks\; and (2) challenging opti
mization landscapes that are common in multi-task learning settings. In bo
th cases\, I will present concrete characterizations of the underlying pro
blems\, and steps we can take to mitigate them. By taking these steps\, we
observe substantial gains in practice.
DTSTART:20200416T163000Z
DTEND:20200416T173000Z
LAST-MODIFIED:20200421T131528Z
LOCATION:Virtual
SUMMARY:Workshop on New Directions in Optimization\, Statistics and Machine
Learning
URL:https://www.ias.edu/node/109991
END:VEVENT
BEGIN:VEVENT
UID:27bebcfa-aef3-4c01-9378-4cdf5ca73f2b
DTSTAMP:20200602T173219Z
CREATED:20200303T185514Z
DESCRIPTION:Topic: Modularity\, Attention and Credit Assignment: Efficient
information dispatching in neural computations\n\nSpeaker: Anirudh Goyal\,
University of Montreal\n\nVideo: https://video.ias.edu/workshop/2020/0415
-16\n\nPhysical processes in the world often have a modular structure\, wi
th complexity emerging through combinations of simpler subsystems. Machine
learning seeks to uncover and use regularities in the physical world. Alt
hough these regularities manifest themselves as statistical dependencies\,
they are ultimately due to dynamic processes governed by physics. These p
rocesses are often independent and only interact sparsely..Despite this\,
most machine learning models employ the opposite inductive bias\, i.e.\, t
hat all processes interact. This can lead to poor generalization (if data
is limited) and lack of robustness to changing task distributions.\n\nIn t
his talk\, I'm going to talk about modularizing the dynamics by learning m
ultiple recurrent modules that are independent by default\, but interact s
paringly\, and show how such an inductive bias gives rise to better out of
distribution generalization.
DTSTART:20200416T180000Z
DTEND:20200416T190000Z
LAST-MODIFIED:20200421T131549Z
LOCATION:Virtual
SUMMARY:Workshop on New Directions in Optimization\, Statistics and Machine
Learning
URL:https://www.ias.edu/node/109936
END:VEVENT
BEGIN:VEVENT
UID:cfc957d0-fff0-4b7a-a97b-60b7c434938a
DTSTAMP:20200602T173219Z
CREATED:20200410T133550Z
DESCRIPTION:Topic: Local-global compatibility in the crystalline case\n\nSp
eaker: Ana Caraiani\, Imperial College\n\nVideo: https://video.ias.edu/pui
as/2020/0416-AnaCaraiani\n\nLet F be a CM field. Scholze constructed Galoi
s representations associated to classes in the cohomology of locally symme
tric spaces for GL_n/F with p-torsion coefficients. These Galois represent
ations are expected to satisfy local-global compatibility at primes above
p. Even the precise formulation of this property is subtle in general\, an
d uses Kisin’s potentially semistable deformation rings. However\, this pr
operty is crucial for proving modularity lifting theorems. I will discuss
joint work with J. Newton\, where we establish local-global compatibility
in the crystalline case under mild technical assumptions. This relies on a
new idea of using P-ordinary parts\, and improves on earlier results obta
ined in joint work with P. Allen\, F. Calegari\, T. Gee\, D. Helm\, B. Le
Hung\, J. Newton\, P. Scholze\, R. Taylor\, and J. Thorne in certain Fonta
ine-Laffaille cases.
DTSTART:20200416T190000Z
DTEND:20200416T200000Z
LAST-MODIFIED:20200416T223227Z
LOCATION:https://theias.zoom.us/j/959183254
SUMMARY:Joint IAS/Princeton University Number Theory Seminar
URL:https://www.ias.edu/node/110441
END:VEVENT
BEGIN:VEVENT
UID:9eda0f9a-9ab2-460b-8d0c-c5c5c0184aac
DTSTAMP:20200602T173219Z
CREATED:20200303T185326Z
DESCRIPTION:Topic: Tradeoffs between Robustness and Accuracy\n\nSpeaker: Pe
rcy Liang\, Stanford University\n\nVideo: https://video.ias.edu/workshop/2
020/0415-16\n\nStandard machine learning produces models that are highly a
ccurate on average but that degrade dramatically when the test distributio
n deviates from the training distribution. While one can train robust mode
ls\, this often comes at the expense of standard accuracy (on the training
distribution). We study this tradeoff in two settings\, adversarial examp
les and minority groups\, creating simple examples which highlight general
ization issues as a major source of this tradeoff. For adversarial example
s\, we show that even augmenting with correct data can produce worse model
s\, but we develop a simple method\, robust self training\, that mitigates
this tradeoff using unlabeled data. For minority groups\, we show that ov
erparametrization of models can also hurt accuracy. These results suggest
that the 'more data' and 'bigger models' strategy that works well for impr
oving standard accuracy need not work on out-of-domain settings\, even in
favorable conditions.
DTSTART:20200416T191500Z
DTEND:20200416T201500Z
LAST-MODIFIED:20200421T131611Z
LOCATION:Virtual
SUMMARY:Workshop on New Directions in Optimization\, Statistics and Machine
Learning
URL:https://www.ias.edu/node/109931
END:VEVENT
BEGIN:VEVENT
UID:4f2f749a-d039-4c5c-ace8-a90655159003
DTSTAMP:20200602T173219Z
CREATED:20200414T171630Z
DESCRIPTION:Topic: Steps towards more human-like learning in machines\n\nSp
eaker: Josh Tenenbaum\, MIT\n\nVideo: https://video.ias.edu/workshop/2020/
0415-16\n\nThere are several broad insights we can draw from computational
models of human cognition in order to build more human-like forms of mach
ine learning. (1) The brain has a great deal of built-in structure\, yet s
till tremendous need and potential for learning. Instead of seeing built-i
n structure and learning as in tension\, we should be thinking about how t
o learn effectively with more and richer forms of structure. (2) The most
powerful forms of human knowledge are symbolic and often causal and probab
ilistic. Symbolic forms allow knowledge to generalize well outside the spa
ce of training data and tasks\, and to be shared publicly and bootstrapped
through collective learning. Probabilistic causal knowledge is actionable
for plans\, supports counterfactual thinking and hypothetical reasoning\,
and allows us to make predictions in situations for which we may have no
relevant data. Human-like machine learning thus has to be able to build sy
mbolic\, causal\, probabilistic representations. I will talk about steps t
owards these goals reflected in some recent work by our group and collabor
ators\, combining techniques from deep learning\, program synthesis\, hier
archical Bayesian modeling and probabilistic programs.
DTSTART:20200416T203000Z
DTEND:20200416T213000Z
LAST-MODIFIED:20200421T131634Z
LOCATION:Virtual
SUMMARY:Workshop on New Directions in Optimization\, Statistics and Machine
Learning
URL:https://www.ias.edu/node/110526
END:VEVENT
BEGIN:VEVENT
UID:8f6907eb-10bb-481f-af94-1b3a200d514d
DTSTAMP:20200602T173219Z
CREATED:20200410T152113Z
DESCRIPTION:Topic: Equivariant quantum operations and relations between the
m\n\nSpeaker: Nicholas Wilkins\, University of Bristol\n\nVideo: https://v
ideo.ias.edu/symplecticgeo/2020/0417-NicholasWilkins\n\nThere is growing i
nterest in looking at operations on quantum cohomology that take into acco
unt symmetries in the holomorphic spheres (such as the quantum Steenrod po
wers\, using a Z/p-symmetry). In order to prove relations between them\, o
ne needs to generalise this to include equivariant operations with more ma
rked points\, varying domains and different symmetry groups. We will look
at the general method of construction of these operations\, as well as two
distinct examples of relations between them.
DTSTART:20200417T131500Z
DTEND:20200417T143000Z
LAST-MODIFIED:20200417T212147Z
LOCATION:https://princeton.zoom.us/j/745635914
SUMMARY:IAS/PU-Montreal-Tel-Aviv Symplectic Geometry Seminar
URL:https://www.ias.edu/node/110451
END:VEVENT
BEGIN:VEVENT
UID:6b8afbdd-529c-4dc1-8e49-a39e42a104de
DTSTAMP:20200602T173219Z
CREATED:20200402T141137Z
DESCRIPTION:Topic: General Discussion\n\n
DTSTART:20200417T190000Z
DTEND:20200417T200000Z
LAST-MODIFIED:20200402T141137Z
LOCATION:Remote Event
SUMMARY:Princeton University HSC Group Discussion
URL:https://www.ias.edu/node/110331
END:VEVENT
BEGIN:VEVENT
UID:cf9c6e42-5e11-412d-8804-3a73a747f903
DTSTAMP:20200602T173219Z
CREATED:20200413T171226Z
DESCRIPTION:Topic: A variational approach to the regularity theory for the
Monge-Ampère equation\n\nSpeaker: Felix Otto\, Max Planck Institute Leipzi
g\n\nVideo: https://video.ias.edu/analysis/2020/0420-FelixOtto\n\nWe prese
nt a purely variational approach to the regularity theory for the Monge-Am
père equation\, or rather optimal transportation\, introduced with M. Gol
dman. Following De Giorgi’s philosophy for the regularity theory of minima
l surfaces\, it is based on the approximation of the displacement by a har
monic gradient\, which leads to a One-Step Improvement Lemma\, and feeds i
nto a Campanato iteration on the $C^{1\,\alpha}$-level for the displacemen
t\, capitalizing on affine invariance. On the one hand\, this allows to re
prove the $C^{1\,\alpha}$-regularity result (Figalli-Kim\, De Philippis-Fi
galli) bypassing Caffarelli’s celebrated theory. This also extends to boun
dary regularity (Chen-Figalli)\, which is joint work with T. Miura. On the
other hand\, due to its robustness\, it can be used as a large-scale regu
larity theory for the problem of matching the Lebesgue measure to the Pois
son measure in the thermodynamic limit.\n\nThis is joint work with M. Gold
man and M. Huesmann.
DTSTART:20200420T150000Z
DTEND:20200420T160000Z
LAST-MODIFIED:20200421T174608Z
LOCATION:https://theias.zoom.us/j/562592856
SUMMARY:Analysis Seminar
URL:https://www.ias.edu/node/110476
END:VEVENT
BEGIN:VEVENT
UID:edcbaa67-96ef-4d62-9f0c-5d81b515e738
DTSTAMP:20200602T173219Z
CREATED:20200415T175351Z
DESCRIPTION:Topic: Structure vs Randomness in Complexity Theory\n\nSpeaker:
Rahul Santhanam\, University of Oxford\n\nVideo: https://video.ias.edu/cs
dm/2020/0420-RahulSanthanam\n\nThe dichotomy between structure and randomn
ess plays an important role in areas such as combinatorics and number theo
ry. I will discuss a similar dichotomy in complexity theory\, and illustra
te it with three examples of my own work: (i) An algorithmic result (with
Igor Oliveira) showing how to probabilistically generate a fixed prime of
length n in time sub-exponential in n\, for infinitely many n (ii) A lower
bound result showing that Promise-MA\, the class of promise problems with
short proofs that are verifiable efficiently by probabilistic algorithms\
, does not have circuits of size n^k for any fixed k (iii) A barrier resul
t (with Jan Pich) showing that certain lower bounds in proof complexity ar
e not themselves efficiently provable.\n\nWhat these results all have in c
ommon is that they are unconditional results in settings where such result
s are often surprising\, and that the proofs are non-constructive. I will
speculate on whether this non-constructivity is essential\, and on the imp
lications for complexity theory more broadly.
DTSTART:20200420T150000Z
DTEND:20200420T160000Z
LAST-MODIFIED:20200421T174647Z
LOCATION:https://theias.zoom.us/j/360043913
SUMMARY:Computer Science/Discrete Mathematics Seminar I
URL:https://www.ias.edu/node/110561
END:VEVENT
BEGIN:VEVENT
UID:2ce2c94f-6c8f-409e-9ffc-355596450550
DTSTAMP:20200602T173219Z
CREATED:20190802T151502Z
DESCRIPTION:Topic: Dark Matter Waves\n\nSpeaker: Lam Hui\, Columbia Univers
ity\n\nA dark matter particle lighter than about 100 eV is expected to exh
ibit wave behavior in a galaxy like the Milky Way. We discuss particle phy
sics motivations for such a scenario\, and implications for astrophysical
observations and experimental detection.
DTSTART:20200420T163000Z
DTEND:20200420T180000Z
LAST-MODIFIED:20200417T133132Z
LOCATION:Until further notice\, meetings will be held remotely on Zoom.
SUMMARY:Institute for Advanced Study/Princeton University Early Universe/Co
smology Lunch Discussion
URL:https://www.ias.edu/node/102641
END:VEVENT
BEGIN:VEVENT
UID:d4771a19-eccd-4337-a7b5-a3dec8466f62
DTSTAMP:20200602T173219Z
CREATED:20200415T175900Z
DESCRIPTION:Topic: Non-commutative optimization: theory\, algorithms and ap
plications (or\, can we prove P!=NP using gradient descent)\n\nSpeaker: A
vi Wigderson\, Herbert H. Maass Professor\, School of Mathematics\n\nVideo
: https://video.ias.edu/csdm/2020/0421-AviWigderson\n\nThis talk aims to s
ummarize a project I was involved in during the past 5 years\, with the ho
pe of explaining our most complete understanding so far\, as well as chall
enges and open problems. The main messages of this project are summarized
below\; I plan to describe\, through examples\, many of the concepts they
refer to\, and the evolution of ideas leading to them. No special backgrou
nd is assumed.\n\n(1) The most basic tools of convex optimization in Eucli
dean space extend to a far more general setting of Riemannian manifolds th
at arise from the symmetries of non-commutative groups. We develop first-o
rder and second-order algorithms\, and analyze their performance in genera
l. While proving convergence bounds requires heavy algebraic and analytic
tools\, convergence itself depends in an elegant way on natural ``smoothne
ss’’ parameters\, in analogy with the Euclidean (commutative) case.\n\n(2)
These algorithms give exponential time (or better) improvements for solvi
ng algorithmic many problems across CS\, Math and Physics. In particular\,
these include problems in algebra (e.g. testing rational identities in no
n-commutative variables)\, in analysis (testing the feasibility and tightn
ess of Brascamp-Lieb inequalities)\, in quantum information theory (to the
quantum marginals problem)\, in algebraic geometry (to computing Kronecke
r coefficients)\, in computational complexity (to derandomizing new specia
l cases of the PIT problem) and in optimization (to testing membership in
large implicitly described polytopes).\n\n(3) The focus on symmetries expo
se old and reveal new relations between the problems above. Essentially\,
they are all membership problems in null cones and moment polytopes of nat
ural group actions on natural spaces. Invariant theory\, which studies suc
h group actions\, plays an essential role in this development. In particul
ar\, a beautiful non-commutative duality theory (expending linear programm
ing duality in the commutative case)\, and notions of geodesic convexity (
extending the Euclidean one) and moment maps (extending the Euclidean grad
ient) are central to the algorithms and their analysis. Interestingly\, mo
st algorithms in invariant theory are symbolic/algebraic\, and these new n
umeric/analytic algorithms proposed here often significntly improve on the
m.\n\nBased on joint works with Zeyuan Allen-Zhu\, Peter Burgisser\, Cole
Franks\, Ankit Garg\, Leonid Gurvits\, Pavel Hrubes\, Yuanzhi Li\, Visu Ma
kam\, Rafael Oliveira and Michael Walter.
DTSTART:20200421T143000Z
DTEND:20200421T163000Z
LAST-MODIFIED:20200421T223631Z
LOCATION:https://theias.zoom.us/j/360043913
SUMMARY:Computer Science/Discrete Mathematics Seminar II
URL:https://www.ias.edu/node/110566
END:VEVENT
BEGIN:VEVENT
UID:ebe4bc07-1473-41c8-8bd4-4694d067df00
DTSTAMP:20200602T173219Z
CREATED:20200415T182658Z
DESCRIPTION:Topic: Assumption-free prediction intervals for black-box regre
ssion algorithms\n\nSpeaker: Aaditya Ramdas\, Carnegie Mellon University\n
\nVideo: https://video.ias.edu/machinelearning/20200421-AadityaRamdas\n\nT
here has been tremendous progress in designing accurate black-box predicti
on methods (boosting\, random forests\, bagging\, neural nets\, etc.) but
for deployment in the real world\, it is useful to quantify uncertainty be
yond making point-predictions. I will summarize recent work that my collab
orators and I have done over the last few years\, for designing a large cl
ass of such methods that enables predictive inference without any assumpti
ons (on the algorithm\, the distribution of the covariates or outcomes)\,
instead just relying on exchangeability of the test and training points.\n
\nI will cover some past work by others\, some recent work by the BaCaRaTi
group (Rina Barber\, Emmanuel Candes\, myself\, Ryan Tibshirani)\, and so
me ongoing work with Arun Kumar Kuchibhotla and Chirag Gupta. [Relevant pa
pers: http://arxiv.org/abs/1905.02928\, https://arxiv.org/pdf/1910.10562.p
df\, https://arxiv.org/abs/1903.04684\, https://arxiv.org/pdf/1904.06019.p
df]
DTSTART:20200421T160000Z
DTEND:20200421T173000Z
LAST-MODIFIED:20200421T223601Z
LOCATION:https://theias.zoom.us/j/384099138
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/110571
END:VEVENT
BEGIN:VEVENT
UID:db888a8d-7805-4877-91f9-9551c75365af
DTSTAMP:20200602T173219Z
CREATED:20200228T182702Z
DESCRIPTION:Topic: Sullivan's Clock: Dennis Sullivan's counter-example to
the periodic orbit conjecture\n\nSpeaker: Richard Schwartz\, Brown Univers
ity\n\nIn 1976 Dennis Sullivan gave an example of a smooth vector-field on
a compact (Riemannian) 5-dimensional manifold in which all the orbits are
closed but for which there is no upper bound to the length of a closed or
bit. (At first this doesn't even seem possible...) Sullivan explained his
proof to me at IHES in 1992 and I still remember his beautiful visual argu
ment - except perhaps for some very fine points. The whole thing boils dow
n to an interesting 'clock-like' isotopy of a smooth loop on the 2-sphere.
In this talk I'll try to convey the construction in non-technical terms a
nd with a lot of pictures and waving of the arms.
DTSTART:20200422T213000Z
DTEND:20200422T230000Z
LAST-MODIFIED:20200420T120237Z
LOCATION:Remote Access Only
SUMMARY:Mathematical Conversations
URL:https://www.ias.edu/node/109816
END:VEVENT
BEGIN:VEVENT
UID:ebce53ca-c0ca-4e7b-ac6d-545ed2abe9de
DTSTAMP:20200602T173219Z
CREATED:20200414T174748Z
DESCRIPTION:Topic: Symmetric power functoriality for holomorphic modular fo
rms\n\nSpeaker: Jack Thorne\, University of Cambridge\n\nLanglands’s funct
oriality conjectures predict the existence of “liftings” of automorphic re
presentations along morphisms of L-groups. A basic case of interest comes
from the irreducible algebraic representations of GL(2)\, thought of as th
e L-group of the reductive group GL(2) over Q. I will discuss the proof\,
joint with James Newton\, of the existence of the corresponding functorial
liftings for a broad class of holomorphic modular forms\, including Raman
ujan’s Delta function
DTSTART:20200423T130000Z
DTEND:20200423T140000Z
LAST-MODIFIED:20200414T174823Z
LOCATION:https://theias.zoom.us/j/959183254
SUMMARY:Joint IAS/Princeton University Number Theory Seminar
URL:https://www.ias.edu/node/110536
END:VEVENT
BEGIN:VEVENT
UID:de89ff75-54bd-4bbe-a15c-bacec2a439ed
DTSTAMP:20200602T173219Z
CREATED:20191025T210001Z
DESCRIPTION:Topic: Deep Generative models and Inverse Problems\n\nSpeaker:
Alexandros Dimakis\, University of Texas at Austin\n\nVideo: https://video
.ias.edu/machinelearning/2020/0423-AlexandrosDimakis\n\nModern deep genera
tive models like GANs\, VAEs and invertible flows are showing amazing resu
lts on modeling high-dimensional distributions\, especially for images. We
will show how they can be used to solve inverse problems by generalizing
compressed sensing beyond sparsity. We will present the general framework\
, new results and open problems in this space.
DTSTART:20200423T190000Z
DTEND:20200423T203000Z
LAST-MODIFIED:20200423T210257Z
LOCATION:https://theias.zoom.us/j/384099138
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/106026
END:VEVENT
BEGIN:VEVENT
UID:6eeacddf-75d5-46f1-b0b9-b428e8ae7412
DTSTAMP:20200602T173219Z
CREATED:20190513T170001Z
DESCRIPTION:
DTSTART:20200424T120000Z
DTEND:20200424T210000Z
LAST-MODIFIED:20200322T195808Z
LOCATION:West Seminar
SUMMARY:CANCELED: NES Hugoye Symposium
URL:https://www.ias.edu/node/100036
END:VEVENT
BEGIN:VEVENT
UID:5a04d4b3-f42f-49d4-89b3-4cc3a42d2d2c
DTSTAMP:20200602T173219Z
CREATED:20190513T170001Z
DESCRIPTION:
DTSTART:20200425T120000Z
DTEND:20200425T210000Z
LAST-MODIFIED:20200322T195808Z
LOCATION:West Seminar
SUMMARY:CANCELED: NES Hugoye Symposium
URL:https://www.ias.edu/node/100036
END:VEVENT
BEGIN:VEVENT
UID:a93254ed-6c13-4293-bcfd-a0b368fcf20d
DTSTAMP:20200602T173219Z
CREATED:20200420T124509Z
DESCRIPTION:Topic: The Geography of Immersed Lagrangian Fillings of Legendr
ian Submanifolds \n\nSpeaker: Lisa Traynor\n\nVideo: https://video.ias.edu
/sympgeo/2020/0424-LisaTraynor\n\nGiven a smooth knot K in the 3-sphere\,
a classic question in knot theory is: What surfaces in the 4-ball have bou
ndary equal to K? One can also consider immersed surfaces and ask a “geogr
aphy” question: What combinations of genus and double points can be realiz
ed by surfaces with boundary equal to K? I will discuss symplectic analogu
es of these questions: Given a Legendrian knot\, what Lagrangian surfaces
can it bound? What immersed Lagrangian surfaces can it bound? These Lagran
gian surfaces are commonly called Lagrangian fillings of the Legendrian kn
ot and are more rigid than their topological counterpart. In particular\,
while any smooth knot bounds an infinite number of topologically distinct
surfaces\, there are classical and non-classical obstructions to the exist
ence of Lagrangian fillings of Legendrian knots. Specifically\, a polynomi
al associated to the Legendrian boundary through the technique of generati
ng families can show that there is no compatible embedded Lagrangian filli
ng. Immersed Lagrangian fillings are more flexible\, and I will describe h
ow this polynomial associated to the Legendrian boundary forbids particula
r combinations of genus and double points in immersed Lagrangian fillings.
In addition\, I will describe some constructions of immersed fillings tha
t allow us to completely answer the Lagrangian geography question for some
Legendrian knots. This is joint work with Samantha Pezzimenti.
DTSTART:20200424T130000Z
DTEND:20200424T143000Z
LAST-MODIFIED:20200424T230356Z
LOCATION:https://princeton.zoom.us/j/745635914
SUMMARY:IAS/PU-Montreal-Tel-Aviv Symplectic Geometry Seminar
URL:https://www.ias.edu/node/110601
END:VEVENT
BEGIN:VEVENT
UID:cb11e952-9a9e-4a96-b129-0683010711fe
DTSTAMP:20200602T173219Z
CREATED:20200402T141207Z
DESCRIPTION:Topic: General Discussion\n\n
DTSTART:20200424T190000Z
DTEND:20200424T200000Z
LAST-MODIFIED:20200402T141207Z
LOCATION:Remote Event
SUMMARY:Princeton University HSC Group Discussion
URL:https://www.ias.edu/node/110336
END:VEVENT
BEGIN:VEVENT
UID:eeeffc69-4d39-4dbd-80de-f662ab17c2a8
DTSTAMP:20200602T173219Z
CREATED:20200219T150300Z
DESCRIPTION:Topic: Graph and Hypergraph Sparsification\n\nSpeaker: Luca Tre
visan\, Bocconi University\n\nVideo: https://video.ias.edu/csdm/2020/0427-
LucaTrevisan\n\nA weighted graph H is a sparsifier of a graph G if H has m
uch fewer edges than G and\, in an appropriate technical sense\, H 'approx
imates' G. Sparsifiers are useful as compressed representations of graphs
and to speed up certain graph algorithms. In a 'cut sparsifier\,' the noti
on of approximation is that every cut is crossed by approximately the same
number of edges in G as in H. In a 'spectral sparsifier' a stronger\, lin
ear-algebraic\, notion of approximation holds. Similar definitions can be
given for hypergraphs. We discuss some new definitions\, new constructions
and new lower bounds for graph and hypergraph sparsifiers\, including: -
A new lower bound on the number edges of any spectral sparsifier\, which c
an be seen as a generalization of the Alon-Boppana theorem to graphs that
are irregular and weighted\; - A new lower bound on the number of bits nee
ded for any data structure that approximately stores the cut information o
f the graph\, showing that known sparsifiers optimally solve this data str
ucture problem - A new definition and constructions of sparsification for
graph and hypergraphs\, which allows *unweighted* sparsifiers with O(n) ed
ges for all graphs and hypergraphs - A new construction of spectral hyperg
raph sparsifiers with a nearly-linear number of hyperedges (compared to a
cubic number of hyperedges in previous constructions).\n\nBased on joint w
ork with Nikhil Bansal\, Charles Carlson\, Alexandra Kolla\, Nikhil Srivas
tava and Ola Svensson.
DTSTART:20200427T150000Z
DTEND:20200427T160000Z
LAST-MODIFIED:20200427T194707Z
LOCATION:https://theias.zoom.us/j/360043913
SUMMARY:Computer Science/Discrete Mathematics Seminar I
URL:https://www.ias.edu/node/109426
END:VEVENT
BEGIN:VEVENT
UID:9d6e4429-a7d0-4b04-a234-02cabd1e5d1a
DTSTAMP:20200602T173219Z
CREATED:20190416T140001Z
DESCRIPTION:Topic: Intensity Mapping with CO and CII as Probes of Primordia
l Non-Gaussianity\n\nSpeaker: Azadeh Moradinezhad\, University of Geneva\n
\nThe non-Gaussianity of the initial conditions leaves its imprints on the
statistical properties of the dark matter and its biased tracers. The mea
surements of the integrated emission of spectral lines from galaxies and t
he intergalactic medium via a technique referred to as line intensity mapp
ing\, provide a mean to probe the large-scale structure at redshifts and s
cales not accessible to traditional galaxy surveys. As such\, line intensi
ty mapping can provide a unique mean to probe primordial non-Gaussianity (
PNG). In this talk\, I will discuss the prospects of constraining intensit
y mapping with rotational lines of carbon monoxide CO\, and the fine struc
ture line of ionised carbon [CII]\, as well as their synergies with galaxy
surveys to constrain PNG. Focusing on two-point statistics\, I will discu
ss the characteristics of optimal survey designs that can achieve the targ
et sensitivity of sigma(fnl) = 1\, as well as the impact of modelling unce
rtainties on the obtained constraints.
DTSTART:20200427T163000Z
DTEND:20200427T180000Z
LAST-MODIFIED:20200422T144541Z
LOCATION:Until further notice\, meetings will be held remotely on Zoom.
SUMMARY:Institute for Advanced Study/Princeton University Early Universe/Co
smology Lunch Discussion
URL:https://www.ias.edu/node/99211
END:VEVENT
BEGIN:VEVENT
UID:d870e243-3c32-4028-b714-161de837aeaa
DTSTAMP:20200602T173219Z
CREATED:20200420T143537Z
DESCRIPTION:Topic: A Framework for Quadratic Form Maximization over Convex
Sets\n\nSpeaker: Vijay Bhattiprolu\, Member\, School of Mathematics\n\nVid
eo: https://video.ias.edu/csdm/2020/0428-VijayBhattiprolu\n\nWe investigat
e the approximability of the following optimization problem\, whose input
is an\nn-by-n matrix A and an origin symmetric convex set C that is given
by a membership oracle:\n'Maximize the quadratic form as x ranges over C.
'\n\nThis is a rich and expressive family of optimization problems\; for d
ifferent choices of forms A\nand convex bodies C it includes a diverse ran
ge of interesting combinatorial and continuous\noptimization problems. To
name some examples\, max-cut\, Grothendieck's inequality\, the\nnon-commut
ative Grothendieck inequality\, certifying hypercontractivity\, small set
expansion\,\nvertex expansion\, and the spread constant of a metric\, are
all captured by this class. While the\nliterature studied these special ca
ses using case-specific reasoning\, here we develop a general\nmethodology
for treatment of the approximability and inapproximability aspects of the
se questions.\n\nBased on joint work with Euiwoong Lee and Assaf Naor.
DTSTART:20200428T143000Z
DTEND:20200428T163000Z
LAST-MODIFIED:20200428T233250Z
LOCATION:https://theias.zoom.us/j/360043913
SUMMARY:Computer Science/Discrete Mathematics Seminar II
URL:https://www.ias.edu/node/110636
END:VEVENT
BEGIN:VEVENT
UID:3bf58756-edb3-4100-8b92-e7dfc006cc40
DTSTAMP:20200602T173219Z
CREATED:20200413T171605Z
DESCRIPTION:Topic: Ellipses of small eccentricity are determined by their D
irichlet (or\, Neumann) spectra\n\nSpeaker: Steven Morris Zelditch \, Nort
hwestern University\n\nVideo: https://video.ias.edu/analysis/2020/0428-Ste
venMorrisZelditch\n\nIn 1965\, M. Kac proved that discs were uniquely dete
rmined by their Dirichlet (or\, Neumann) spectra. Until recently\, disks w
ere the only smooth plane domains known to be determined by their eigenval
ues. Recently\, H. Hezari and I proved that ellipses of small eccentricity
are also determined uniquely by their Dirichlet (or\, Neumann) spectra. T
he proof uses recent results of Avila\, de Simoi\, and Kaloshin\, proving
that nearly circular plane domains with rationally integrable billiards mu
st be ellipses. It also uses a ``bounce decomposition'' for the wave trace
\, representing the wave trace as a sum of q-bounce oscillatory integrals.
It is shown that for nearly circular domains\, each is a spectral invaria
nt and that the ellipse is uniquely determined by its q-bounce invariants.
DTSTART:20200428T150000Z
DTEND:20200428T160000Z
LAST-MODIFIED:20200428T233345Z
LOCATION:https://theias.zoom.us/j/562592856
SUMMARY:Analysis Seminar
URL:https://www.ias.edu/node/110481
END:VEVENT
BEGIN:VEVENT
UID:38740dac-6450-4985-8da6-2c7315e9bbcf
DTSTAMP:20200602T173219Z
CREATED:20200420T120412Z
DESCRIPTION:Topic: Musings about a 10-year collaboration with biological mo
rphologists\, or how to make biologists comfortable with fiber bundles.\n
\nSpeaker: Ingrid Daubechies\, Duke University\n\nIn the course of this co
llaboration\, both sides learned about the other field\; to my surprise\,
the biologists learned to 'speak' some mathematics. Also\, when they saw h
ow we approached answering their initial questions\, the questions changed
. And finally\, the collaboration led to interesting mathematical question
s.
DTSTART:20200429T213000Z
DTEND:20200429T230000Z
LAST-MODIFIED:20200424T190129Z
LOCATION:Remote Access Only
SUMMARY:Mathematical Conversations
URL:https://www.ias.edu/node/110581
END:VEVENT
BEGIN:VEVENT
UID:afdb5b3e-cbb4-4e8b-a290-01ab6a65f2d6
DTSTAMP:20200602T173219Z
CREATED:20191025T210001Z
DESCRIPTION:Topic: Latent Stochastic Differential Equations for Irregularly
-Sampled Time Series\n\nSpeaker: David Duvenaud\, University of Toronto\n
\nMuch real-world data is sampled at irregular intervals\, but most time s
eries models require regularly-sampled data. Continuous-time models addres
s this problem\, but until now only deterministic (ODE) models or linear-G
aussian models were efficiently trainable with millions of parameters. We
construct a scalable algorithm for computing gradients of samples from sto
chastic differential equations (SDEs)\, and for gradient-based stochastic
variational inference in function space\, all with the use of adaptive bla
ck-box SDE solvers. This allows us to fit a new family of richly-parameter
ized distributions over time series. We apply latent SDEs to motion captur
e data\, and will also discuss prototypes of infinitely-deep Bayesian neur
al networks.
DTSTART:20200430T190000Z
DTEND:20200430T203000Z
LAST-MODIFIED:20200424T184212Z
LOCATION:Remote Access Only - see link below
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/106031
END:VEVENT
BEGIN:VEVENT
UID:3218f136-2e5b-43b8-a68a-753f39d0941f
DTSTAMP:20200602T173219Z
CREATED:20200413T172110Z
DESCRIPTION:Topic: Eulerianity of Fourier coefficients of automorphic forms
\n\nSpeaker: Henrik Gustafsson\, Member\, School of Mathematics\n\nThe fac
torization of Fourier coefficients of automorphic forms plays an important
role in a wide range of topics\, from the study of L-functions to the int
erpretation of scattering amplitudes in string theory. In this talk I will
present a transfer theorem which derives the Eulerianity of certain Fouri
er coefficients from that of another coefficient. I will also discuss some
applications of this theorem to Fourier coefficients of automorphic forms
in minimal and next-to-minimal representations.\n\nBased on recent work w
ith Dmitry Gourevitch\, Axel Kleinschmidt\, Daniel Persson and Siddhartha
Sahi.
DTSTART:20200430T203000Z
DTEND:20200430T213000Z
LAST-MODIFIED:20200424T121944Z
LOCATION:https://theias.zoom.us/j/959183254
SUMMARY:Joint IAS/Princeton University Number Theory Seminar
URL:https://www.ias.edu/node/110486
END:VEVENT
BEGIN:VEVENT
UID:c5fe70dc-2a12-42cd-8fde-eb330608a0c1
DTSTAMP:20200602T173219Z
CREATED:20200414T173853Z
DESCRIPTION:Topic: Zoll contact forms are local maximisers of the systolic
ratio\n\nSpeaker: Alberto Abbondandolo\, Ruhr-Universität Bochum \n\nA cen
tral question from systolic geometry is to find upper bounds for the systo
lic ratio of a Riemannian metric on a closed $n$-dimensional manifold\, i.
e. the ratio of the $n$-th power of the shortest length of closed geodesic
s by the volume. This question can be naturally extended to Reeb flows\, a
class of dynamical systems including geodesic flows and induced by a cont
act form on a closed manifold. The aim of this talk is to discuss a recent
result obtained in collaboration with Gabriele Benedetti: Zoll contact fo
rms\, i.e. forms such that all the orbits of the induced Reeb flow are per
iodic with the same period\, are local maximisers of the systolic ratio. C
onsequences of this result are: (i) sharp systolic inequalities for Rieman
nian and Finsler metrics close to Zoll ones\, (ii) the perturbative case o
f a conjecture of Viterbo on the symplectic capacity of convex bodies\, (i
ii) a generalization of Gromov's non-squeezing theorem in the intermediate
dimensions for symplectomorphisms that are close to linear ones.
DTSTART:20200501T130000Z
DTEND:20200501T143000Z
LAST-MODIFIED:20200423T160502Z
LOCATION:https://princeton.zoom.us/j/745635914
SUMMARY:IAS/PU-Montreal-Tel-Aviv Symplectic Geometry Seminar
URL:https://www.ias.edu/node/110531
END:VEVENT
BEGIN:VEVENT
UID:e5ecd56b-991c-4e2b-bb7b-201a18f87775
DTSTAMP:20200602T173219Z
CREATED:20200128T185518Z
DESCRIPTION:Topic: General Discussion\n\n
DTSTART:20200501T190000Z
DTEND:20200501T200000Z
LAST-MODIFIED:20200424T142601Z
LOCATION:Remote Event
SUMMARY:Princeton University HSC Group Discussion
URL:https://www.ias.edu/node/108311
END:VEVENT
BEGIN:VEVENT
UID:29eb5e6e-f69d-4372-8f93-e024b8b868b3
DTSTAMP:20200602T173219Z
CREATED:20200128T192314Z
DESCRIPTION:Topic: Stellar Mass-metallicity Relations at z ~ 2-3\n\nSpeaker
: Discussion Leader Allison Strom\, Carnegie Observatories\n\n
DTSTART:20200504T150000Z
DTEND:20200504T160000Z
LAST-MODIFIED:20200501T135937Z
LOCATION:Remote via Zoom
SUMMARY:Princeton University Galread [Galactic/Extragalactic Reading Group]
URL:https://www.ias.edu/node/108466
END:VEVENT
BEGIN:VEVENT
UID:1f31ad96-4325-4b00-936b-23663051c335
DTSTAMP:20200602T173219Z
CREATED:20200413T173744Z
DESCRIPTION:Topic: Exponential mixing of 3D Anosov flows\n\nSpeaker: Zhiyua
n Zhang\, Université Paris 13\n\nVideo: https://video.ias.edu/analysis/202
0/0504-ZhiyuanZhang\n\nWe show that a topologically mixing C^\infty Anosov
flow on a 3 dimensional compact manifold is exponential mixing with respe
ct to any equilibrium measure with Holder potential. This is a joint work
with Masato Tsujii.
DTSTART:20200504T150000Z
DTEND:20200504T160000Z
LAST-MODIFIED:20200505T140127Z
LOCATION:https://theias.zoom.us/j/562592856
SUMMARY:Analysis Seminar
URL:https://www.ias.edu/node/110491
END:VEVENT
BEGIN:VEVENT
UID:731ec822-073f-4e8c-b58b-16597af84989
DTSTAMP:20200602T173219Z
CREATED:20200423T124043Z
DESCRIPTION:Topic: Local Statistics\, Semidefinite Programming\, and Commun
ity Detection\n\nSpeaker: Prasad Raghavendra\, University of California\,
Berkeley\n\nVideo: https://video.ias.edu/csdm/2020/0504-PrasadRaghavendra
\n\nWe propose a new hierarchy of semidefinite programming relaxations for
inference problems. As test cases\, we consider the problem of community
detection in block models. The vertices are partitioned into k communities
\, and a graph is sampled conditional on a prescribed number of inter- and
intra-community edges. The problem of detection\, where we are to decide
with high probability whether a graph was drawn from this model or the uni
form distribution on regular graphs\, is conjectured to undergo a computat
ional phase transition at a point called the Kesten-Stigum (KS) threshold.
We consider two models of random graphs namely the well-studied (irregula
r) stochastic block model and a distribution over random regular graphs we
'll call the degree-regular block model (DRBM). For both these models\, we
show that sufficiently high constant levels of our hierarchy can perform
detection arbitrarily close to the KS threshold and that our algorithm is
robust to up to a linear number of adversarial edge perturbations. Further
more\, in the case of DRBM\, we show that below the Kesten-Stigum threshol
d no constant level can do so.\n\nJoint work with Jess Banks (U.C.Berkeley
) and Sidhanth Mohanty (U.C. Berkeley)
DTSTART:20200504T150000Z
DTEND:20200504T160000Z
LAST-MODIFIED:20200505T140304Z
LOCATION:https://theias.zoom.us/j/360043913
SUMMARY:Computer Science/Discrete Mathematics Seminar I
URL:https://www.ias.edu/node/110681
END:VEVENT
BEGIN:VEVENT
UID:e9a8c2ed-4700-4297-8481-a9ea31b7b2b3
DTSTAMP:20200602T173219Z
CREATED:20200123T193714Z
DESCRIPTION:Topic: A Giant Planet Candidate Transiting a White Dwarf\n\nSpe
aker: Andrew Vanderburg\, University of Texas at Austin\n\nAstronomers hav
e discovered thousands of planets outside the solar system\, most of which
orbit stars that will eventually evolve into red giants and then into whi
te dwarfs. During the red giant phase\, any close-orbiting planets will be
engulfed by the star\, but more distant planets can survive this phase an
d remain in orbit around the white dwarf. Some white dwarfs show evidence
for rocky material floating in their atmospheres\, in warm debris disks\,
or orbiting very closely\, which has been interpreted as the debris of roc
ky planets that were scattered inward and tidally disrupted. Recently\, th
e discovery of a gaseous debris disk with a composition similar to ice gia
nt planets demonstrated that massive planets might also find their way int
o tight orbits around white dwarfs\, but it is unclear whether the planets
can survive the journey. So far\, the detection of intact planets in clos
e orbits around white dwarfs has remained elusive. Here\, we report the di
scovery of a giant planet candidate transiting the white dwarf WD 1856+534
(TIC 267574918) every 1.4 days. The planet candidate is roughly the same
size as Jupiter and is no more than 14 times as massive as Jupiter (with 9
5% confidence). Other cases of white dwarfs with close brown dwarf or stel
lar companions are explained as the consequence of common-envelope evoluti
on\, wherein the original orbit is enveloped during the red-giant phase an
d shrinks due to friction. In this case\, though\, the low mass and relati
vely long orbital period of the planet candidate make this scenario unlike
ly. Instead\, the WD 1856+534 system seems to demonstrate that giant plane
ts can be scattered into tight orbits without being tidally disrupted\, an
d motivates searches for smaller transiting planets around white dwarfs.
DTSTART:20200504T161500Z
DTEND:20200504T171500Z
LAST-MODIFIED:20200501T141536Z
LOCATION:Virtual Meeting
SUMMARY:Princeton University Exoplanet Discussion Group
URL:https://www.ias.edu/node/108046
END:VEVENT
BEGIN:VEVENT
UID:f96514b3-822f-4efa-a1a1-bd628161256e
DTSTAMP:20200602T173219Z
CREATED:20190802T151502Z
DESCRIPTION:Topic: A new halo model formalism for the cosmic infrared backg
round (CIB) and its correlation with the thermal Sunyaev-Zel’dovich effect
(tSZ)\n\nSpeaker: Abhishek Maniyar\, New York University\n\nThe distinct
frequency-redshift dependence of the CIB allows to probe a large span of r
edshifts. In the last few years\, anisotropies of the CIB have become an i
mportant cosmological tool as a tracer of the large-scale structures. Howe
ver\, on very small angular scales\, the CIB along with the tSZ act as for
egrounds while measuring the kinetic SZ effect from the CMB power spectrum
. The CIB is expected to be correlated with the tSZ and this CIB-tSZ corre
lation further acts as a foreground as well to the kSZ measurements. Thus
it is important to have physically motivated models for these foregrounds
to properly account for them. I will present a new halo model for the CIB
which is simplistic in nature with four parameters and is able to fit both
the Planck as well as Herschel CIB power spectra measurements. We use thi
s new halo model for the CIB along with a halo model for the tSZ to calcul
ate the CIB-tSZ correlation.
DTSTART:20200504T163000Z
DTEND:20200504T180000Z
LAST-MODIFIED:20200501T141843Z
LOCATION:Until further notice\, the meetings will be held remotely on Zoom.
SUMMARY:Institute for Advanced Study/Princeton University Early Universe/Co
smology Lunch Discussion
URL:https://www.ias.edu/node/102646
END:VEVENT
BEGIN:VEVENT
UID:2fdc1fe6-ab03-4013-bb7b-7cd60c9de7ab
DTSTAMP:20200602T173219Z
CREATED:20200501T170824Z
DESCRIPTION:Topic: Data Spring Cleaning: Data File Inventory\n\nPlease see
the link below from the Princeton University Library as they will be offer
ing the Princeton Research Data Service Discussion Series which you may fi
nd of interest. The first session starts this coming Monday\, May 4. Here
is the link to register: [https://libcal.princeton.edu/calendar/?cid=12260
&t=d&d=0000-00-00&cal=12260&ct=49413.](https://libcal.princeton.edu/calend
ar/?cid=12260&t=d&d=0000-00-00&cal=12260&ct=49413) Please direct your ques
tions to Emily Judd\, at the Princeton University Library.
DTSTART:20200504T170000Z
DTEND:20200504T180000Z
LAST-MODIFIED:20200501T170824Z
LOCATION:Virtual Event
SUMMARY:Princeton University Research Data Service Discussion Series
URL:https://www.ias.edu/node/110791
END:VEVENT
BEGIN:VEVENT
UID:8088cc08-7189-45ec-b699-a8cb8e13667a
DTSTAMP:20200602T173219Z
CREATED:20200501T184127Z
DESCRIPTION:Topic: Data Spring Cleaning: Data File Inventory\n\nJoin the di
scussion and exchange tips on how to keep track of your data files more ef
ficiently and save yourself headaches down the road. A zoom link will be p
rovided by email. This event is part of Princeton Research Data Service Di
scussion Series\, Data Spring Cleaning: Or\, How Can We Help Your Data “Sp
ark Joy”? Here is the link to register: https://libcal.princeton.edu/calen
dar/?cid=12260&t=d&d=0000-00-00&cal=12260&ct=49413 . Please direct your qu
estions to Emily Judd (ejudd AT princeton.edu) at the Princeton University
Library.
DTSTART:20200504T170000Z
DTEND:20200504T180000Z
LAST-MODIFIED:20200501T184556Z
LOCATION:Virtual
SUMMARY:Princeton University Research Data Service Discussion Series
URL:https://www.ias.edu/node/110801
END:VEVENT
BEGIN:VEVENT
UID:084000d8-a4d1-4b18-b815-fcda3ddd3faf
DTSTAMP:20200602T173219Z
CREATED:20200423T184154Z
DESCRIPTION:Topic: Recent Progress on Cutting Planes Proofs\n\nSpeaker: Noa
h Fleming\, University of Toronto\n\nVideo: https://video.ias.edu/csdm/202
0/0505-NoahFleming\n\nProof Complexity studies the length of proofs of pro
positional tautologies in various restricted proof systems. One of the mos
t well-studied is the Cutting Planes proof system\, which captures reasoni
ng which can be expressed using linear inequalities. A series of papers pr
oved lower bounds on the length of Cutting Planes using the method of feas
ible interpolation whereby proving lower bounds on the size of Cutting Pla
nes lower bounds proofs of a certain restricted class of formulas is reduc
ed to monotone circuit lower bounds. However\, it was not known how to obt
ain lower bounds for more natural formulas such as the Tseitin (mod 2) for
mulas or random formulas.\n\nIn this talk I will discuss two recent result
s for Cutting Planes. The first is joint work together with Pankratov\, Pi
tassi\, and Robere (and proved independently by Hrubeš and Pudlak) which p
roves exponential lower bounds on the length of Cutting Planes proofs of r
andom Θ(log n)-CNF formulas. The second is a surprising result of Dadush a
nd Tiwari which gives quasi-polynomial size Cutting Planes proofs of the T
seitin tautologies. My aim is to highlight the main conceptual ideas in th
e proofs of both of these results.
DTSTART:20200505T143000Z
DTEND:20200505T163000Z
LAST-MODIFIED:20200505T140404Z
LOCATION:https://theias.zoom.us/j/360043913
SUMMARY:Computer Science/Discrete Mathematics Seminar II
URL:https://www.ias.edu/node/110741
END:VEVENT
BEGIN:VEVENT
UID:de7433f3-14d6-4b97-8a79-7910966d2287
DTSTAMP:20200602T173219Z
CREATED:20200423T125306Z
DESCRIPTION:Topic: Boosting Simple Learners\n\nSpeaker: Shay Moran\, Google
\n\nVideo: https://video.ias.edu/machinelearning/2020/0505-ShayMoran\n\nWe
study boosting algorithms under the assumption that the given weak learne
r outputs hypotheses from a class of bounded capacity. This assumption is
inspired by the common convention that weak hypotheses are “rules-of-thumb
s” from an “easy-to-learn class”. Formally\, we assume the class of weak h
ypotheses has a bounded VC dimension.\n\nWe focus on two main questions:\n
\n(i) Oracle Complexity: we show that this restriction allows to circumven
t a lower bound due to Freund and Schapire (‘95\, ‘12) on the number of ca
lls to the weak learner. Unlike previous boosting algorithms which aggrega
te the weak hypotheses by majority votes\, in order to circumvent the lowe
r bound we propose a boosting procedure that uses more complex aggregation
rules\, and we show this to be necessary.\n\n(ii) Expressivity: Which tas
ks can be learned by boosting a base-class of bounded VC-dimension? Can co
ncepts that are “far away” from the class be learned? Towards answering th
is question we identify a combinatorial-geometric parameter which quantifi
es the expressivity of the base-class when used as part of a boosting proc
edure. Along the way\, we establish and exploit connections with Discrepan
cy Theory.\n\nJoint with Noga Alon\, Alon Gonen\, and Elad Hazan
DTSTART:20200505T160000Z
DTEND:20200505T173000Z
LAST-MODIFIED:20200505T140341Z
LOCATION:Remote Access Only - see link below
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/110691
END:VEVENT
BEGIN:VEVENT
UID:b3158103-262b-433e-8e0a-2b45db8e875c
DTSTAMP:20200602T173219Z
CREATED:20200501T184937Z
DESCRIPTION:Topic: Data Spring Cleaning: Tidy Data and Documentation\n\nJoi
n the discussion and exchange tips on on how to tidy up your datasets and
maintain consistent documentation through all of your projects. A zoom lin
k will be provided by email. This event is part of Princeton Research Data
Service Discussion Series\, Data Spring Cleaning: Or\, How Can We Help Yo
ur Data “Spark Joy”? Here is the link to register: https://libcal.princeto
n.edu/calendar/?cid=12260&t=d&d=0000-00-00&cal=12260&ct=49413 . Please dir
ect your questions to Emily Judd (ejudd AT princeton.edu) at the Princeton
University Library.
DTSTART:20200506T170000Z
DTEND:20200506T180000Z
LAST-MODIFIED:20200501T184937Z
LOCATION:Virtual
SUMMARY:Princeton University Research Data Service Discussion Series
URL:https://www.ias.edu/node/110806
END:VEVENT
BEGIN:VEVENT
UID:47e4d5fb-7612-48d5-ae89-bafae9bfd723
DTSTAMP:20200602T173219Z
CREATED:20200424T172833Z
DESCRIPTION:Topic: Discrepancy Theory and Randomized Controlled Trials\n\nS
peaker: Daniel Spielman\, Yale University\n\nDiscrepancy theory tells us t
hat it is possible to partition vectors into sets so that each set looks s
urprisingly similar to every other. By 'surprisingly similar' we mean much
more similar than a random partition. Randomized Controlled Trials are us
ed to test the effectiveness of interventions\, like medical treatments. R
andomization is used to ensure that the test and control groups are probab
ly similar. When we know nothing about the experimental subjects\, uniform
random assignment is the best we can do. When we have information about t
he experimental subjects that is likely to be correlated with medical outc
omes\, we can try to combine the strengths of randomization with the promi
ses of discrepancy theory. This should allow us to obtain more accurate es
timates of the effectiveness of treatments\, or to conduct trials with few
er experimental subjects. This intersection raises many exciting problems
in discrepancy theory and experimental design.\n\nOn Monday\, May 11\, I w
ill present a paper on this topic that Christopher Harshaw\, Fredrik Sävje
\, and Peng Zhang and I are finishing writing.
DTSTART:20200506T213000Z
DTEND:20200506T230000Z
LAST-MODIFIED:20200505T123637Z
LOCATION:Remote Access Only
SUMMARY:Mathematical Conversations
URL:https://www.ias.edu/node/110781
END:VEVENT
BEGIN:VEVENT
UID:c818c760-5ed9-4de1-bf98-9a7451e669e7
DTSTAMP:20200602T173219Z
CREATED:20200501T191641Z
DESCRIPTION:
DTSTART:20200507T163000Z
DTEND:20200507T173000Z
LAST-MODIFIED:20200504T160341Z
LOCATION:
SUMMARY:Princeton University Thunch Talk - Postponed
URL:https://www.ias.edu/node/110816
END:VEVENT
BEGIN:VEVENT
UID:84915e02-ba6a-473f-888c-1b01d0196ca8
DTSTAMP:20200602T173219Z
CREATED:20191025T211502Z
DESCRIPTION:Topic: Learning probability distributions\; What can\, What can
't be done\n\nSpeaker: Shai Ben-David\, University of Waterloo\n\nA possib
le high level description of statistical learning is that it aims to learn
about some unknown probability distribution ('environment”) from samples
it generates ('training data”). In its most general form\, assuming no pri
or knowledge and asking to find accurate approximations to the data genera
ting distributions\, there can be no success guarantee. In this talk I wil
l discuss two major directions of relaxing that too hard problem.\n\nFirst
\, I will address the situation under common prior knowledge assumption -
I will describe settling the question of the sample complexity of learning
mixtures of Gaussians.\n\nSecondly\, I will address what can be learnt ab
out unknown distributions when no prior knowledge is applied. I will descr
ibe a surprising result. Namely\, the independence from set theory of a ba
sic statistical learnability problem. As a corollary\, I will show that th
ere can be no combinatorial dimension that characterizes the families of r
andom variables that can be reliably learnt (in contrast with the known VC
-dimension like characterizations of common supervised learning tasks).\n
\nBoth parts of the talks use novel notions of sample compression schemes
as key components.\n\nThe first part is based on joint work with Hasan Ash
iani\, Nick Harvey\, Chris Law\, Abas Merhabian and Yaniv Plan and the sec
ond part on work with Shay Moran\, Pavel Hrubes\, Amir Shpilka and Amir Ye
hudayoff.
DTSTART:20200507T190000Z
DTEND:20200507T203000Z
LAST-MODIFIED:20200430T173545Z
LOCATION:Remote Access Only - see link below
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/106036
END:VEVENT
BEGIN:VEVENT
UID:b28a9e82-c620-481f-9a3c-b634a5d8ffaf
DTSTAMP:20200602T173219Z
CREATED:20200420T122550Z
DESCRIPTION:Topic: On triple product L functions\n\nSpeaker: Jayce Robert G
etz\, Duke University\n\nEstablishing the conjectured analytic properties
of triple product L-functions is a crucial case of Langlands functoriality
. However\, little is known. I will present work in progress on the case o
f triples of automorphic representations on GL_3\; in some sense this is t
he smallest case that appears out of reach via standard techniques. The ap
proach is based on the beautiful fibration method of Braverman and Kazhdan
for constructing Schwartz spaces and proving analogues of the Poisson sum
mation formula.
DTSTART:20200507T203000Z
DTEND:20200507T213000Z
LAST-MODIFIED:20200430T184222Z
LOCATION:https://theias.zoom.us/j/959183254
SUMMARY:Joint IAS/Princeton University Number Theory Seminar
URL:https://www.ias.edu/node/110591
END:VEVENT
BEGIN:VEVENT
UID:36a2d3f9-e60a-4de0-a04b-5667f43b0257
DTSTAMP:20200602T173219Z
CREATED:20200420T142455Z
DESCRIPTION:Topic: Spectral characterizations of Besse and Zoll Reeb flows
\n\nSpeaker: Marco Mazzucchelli\, École normale supérieure de Lyon\n\nVide
o: https://video.ias.edu/symplectic/2020/0508-MarcoMazzucchelli\n\nIn this
talk\, I will address a geometric inverse problem from contact geometry:
is it possible to recognize whether all orbits of a given Reeb flow are cl
osed from the knowledge of the action spectrum? Borrowing the terminology
from Riemannian geometry\, Reeb flows all of whose orbits are closed are s
ometimes called Besse\, and Besse Reeb flows all of whose orbits have the
same minimal period are sometimes called Zoll. In the talk I will summariz
e recent results on this inverse problem in a few settings: geodesic flows
(joint work with Stefan Suhr)\, closed contact 3-manifolds (joint work wi
th Daniel Cristofaro-Gardiner)\, convex contact spheres and\, more general
ly\, restricted contact type hypersurfaces of symplectic vector spaces (jo
int work with Viktor Ginzburg and Basak Gürel). I will also mention a few
conjectures and open problems.
DTSTART:20200508T130000Z
DTEND:20200508T143000Z
LAST-MODIFIED:20200515T011613Z
LOCATION:https://princeton.zoom.us/j/745635914
SUMMARY:IAS/PU-Montreal-Tel-Aviv Symplectic Geometry Seminar
URL:https://www.ias.edu/node/110616
END:VEVENT
BEGIN:VEVENT
UID:657999bd-952a-4ef1-a4d6-2f0b7bc8b322
DTSTAMP:20200602T173219Z
CREATED:20200501T185326Z
DESCRIPTION:Topic: Data Spring Cleaning: Data Shelves\, Closets\, and Attic
s\n\nJoin the discussion and exchange tips on how to better organize your
active data files and keep your analysis workspace free from clutter. A zo
om link will be provided by email. This event is part of Princeton Researc
h Data Service Discussion Series\, Data Spring Cleaning: Or\, How Can We H
elp Your Data “Spark Joy”? Here is the link to register: https://libcal.pr
inceton.edu/calendar/?cid=12260&t=d&d=0000-00-00&cal=12260&ct=49413 . Plea
se direct your questions to Emily Judd (ejudd AT princeton.edu) at the Pri
nceton University Library.
DTSTART:20200508T170000Z
DTEND:20200508T180000Z
LAST-MODIFIED:20200501T190034Z
LOCATION:Virtual
SUMMARY:Princeton University Research Data Service Discussion Series
URL:https://www.ias.edu/node/110811
END:VEVENT
BEGIN:VEVENT
UID:5af9784f-d6bd-4932-b0d4-e8a12abe6557
DTSTAMP:20200602T173219Z
CREATED:20200128T185559Z
DESCRIPTION:Topic: General Discussion\n\n
DTSTART:20200508T190000Z
DTEND:20200508T200000Z
LAST-MODIFIED:20200501T142007Z
LOCATION:Remote Event
SUMMARY:Princeton University HSC Group Discussion
URL:https://www.ias.edu/node/108316
END:VEVENT
BEGIN:VEVENT
UID:47200977-d6e5-44b7-8f3c-97c46ae43d38
DTSTAMP:20200602T173219Z
CREATED:20200124T160538Z
DESCRIPTION:Topic: Using discrepancy theory to improve the design of random
ized controlled trials\n\nSpeaker: Daniel Spielman\, Yale University\n\nVi
deo: https://video.ias.edu/csdm/2020/0511-DanielSpielman\n\nIn randomized
experiments\, such as a medical trials\, we randomly assign the treatment\
, such as a drug or a placebo\, that each experimental subject receives. R
andomization can help us accurately estimate the difference in treatment e
ffects with high probability. We also know that we want the two groups to
be similar: ideally the two groups would be similar in every statistic we
can measure beforehand. Recent advances in algorithmic discrepancy theory
allow us to divide subjects into groups with similar statistics.\n\nBy exp
loiting the Gram-Schmidt Walk algorithm of Bansal\, Dadush\, Garg\, and Lo
vett\, we can obtain random assignments of low discrepancy. These allow us
to obtain more accurate estimates of treatment effects when the informati
on we measure about the subjects is predictive\, while also bounding the w
orst-case behavior when it is not.\n\nIn this talk\, I will formally expla
in the problem of estimating treatment effects in randomized controlled tr
ials\, the dangers of using fancy inference techniques instead of fancy de
signs\, how we use the Gram-Schmidt Walk algorithm\, a tight analysis of t
his algorithm\, and how we use it to obtain confidence intervals. I hope t
o explain just how much we don't yet know.\n\nThis is joint work with Chri
stopher Harshaw\, Fredrik Sävje\, and Peng Zhang.
DTSTART:20200511T150000Z
DTEND:20200511T160000Z
LAST-MODIFIED:20200515T011654Z
LOCATION:https://theias.zoom.us/j/360043913
SUMMARY:Computer Science/Discrete Mathematics Seminar I
URL:https://www.ias.edu/node/108111
END:VEVENT
BEGIN:VEVENT
UID:873e427b-5787-445e-9948-cee645a4b1b5
DTSTAMP:20200602T173219Z
CREATED:20200508T140116Z
DESCRIPTION:Topic: Tilting Planets and Keeping Them There: Generating Plane
tary Obliquities with Cassini State Resonances\n\nSpeaker: Yubo Su\, Corne
ll University\n\nIn most scenarios of planet formation\, planets form out
of a protoplanetary disk with a dominant angular momentum axis. As such\,
planetary obliquities\, the misalignment angle between the spin and orbita
l axis of a planet\, are naively expected to be small. Yet in our Solar Sy
stem\, the planets' obliquities are substantial\, notably those of Jupiter
(3.12°) and Saturn (26.73°). These two obliquities are thought to be the
result of the giant planets being in a particular secular resonance called
a Cassini state. In this talk\, I will discuss two dynamical mechanisms b
y which substantial planetary obliquities can be generated and maintained
in exoplanetary systems by this same Cassini state resonance.
DTSTART:20200511T161500Z
DTEND:20200511T171500Z
LAST-MODIFIED:20200508T140530Z
LOCATION:Virtual Meeting
SUMMARY:Princeton University Exoplanet Discussion Group
URL:https://www.ias.edu/node/110921
END:VEVENT
BEGIN:VEVENT
UID:84b2d660-9b78-45e9-8978-aed3c048b467
DTSTAMP:20200602T173219Z
CREATED:20190416T134501Z
DESCRIPTION:Topic: What Can the Frontiers of Clusters Teach Us About the Da
rk Sector?\n\nSpeaker: Omar Contigiani\, Leiden University\n\nIn N-body si
mulations\, the splashback radius marks the boundary of massive haloes\, w
here infalling and already collapsed material meet. The associated feature
is a direct probe of the growth of structure\, and its dynamics are relat
ively simple. I will discuss what this feature is\, how it has been measur
ed\, and what it can tell us about dark matter and dark energy.
DTSTART:20200511T163000Z
DTEND:20200511T180000Z
LAST-MODIFIED:20200508T133630Z
LOCATION:Until further notice\, meetings will be held remotely on Zoom.
SUMMARY:Institute for Advanced Study/Princeton University Early Universe/Co
smology Lunch Discussion
URL:https://www.ias.edu/node/99161
END:VEVENT
BEGIN:VEVENT
UID:3f0e0fec-ddf5-4a59-bcd8-86b979771124
DTSTAMP:20200602T173219Z
CREATED:20200423T184346Z
DESCRIPTION:Topic: Convex Set Disjointness\, Distributed Learning of Halfsp
aces\, and Linear Programming\n\nSpeaker: Shay Moran\, Member\, School of
Mathematics\n\nVideo: https://video.ias.edu/csdm/2020/0512-ShayMoran\n\nDi
stributed learning protocols are designed to train on distributed data wit
hout gathering it all on a single centralized machine\, thus contributing
to the efficiency of the system and enhancing its privacy. We study a cent
ral problem in distributed learning\, called Distributed Learning of Halfs
paces: let U \subset R^d be a known domain of size n and let h:R^d —> R be
an unknown target affine function. A set of examples {(u\,b)} is distribu
ted between several parties\, where u \in U is a point and b = sign(h(u))
\in {-1\, +1} is its label. The parties' goal is to agree\, using minimum
communication\, on a classifier f: U—>{-1\,+1} such that f(u)=b for every
input example (u\,b). (In practice\, the finite domain U is defined implic
itly by the representation of d-dimensional vectors which is used in the p
rotocol.) We establish a (nearly) tight bound of ~$\Theta$ (d*log n) on th
e communication complexity of the problem of distributed learning of halfs
paces in the two-party setting. Since this problem is closely related to t
he Convex Set Disjointness problem in communication complexity and the pro
blem of Distributed Linear Programming in distributed optimization\, we ar
e able to derive upper and lower bounds of ~O(d^2\log n) and ~Ω(d\log n) f
or both of these basic problems as well. Our main technical contribution l
ies in the design and analysis of the protocols which imply the upper boun
ds. To this end\, we introduce a technique called Halfspace Containers\, a
llowing for a compressed approximate representation of every halfspace. Ha
lfspace containers may be of independent interest and are closely related
to bracketing numbers in statistics and to hyperplane cuttings in discrete
geometry. Joint paper with Mark Braverman\, Gillat Kol\, and Raghuvansh R
Saxena
DTSTART:20200512T143000Z
DTEND:20200512T163000Z
LAST-MODIFIED:20200515T011733Z
LOCATION:https://theias.zoom.us/j/360043913
SUMMARY:Computer Science/Discrete Mathematics Seminar II
URL:https://www.ias.edu/node/110746
END:VEVENT
BEGIN:VEVENT
UID:96b3ba0d-5047-4c4b-bd8f-ba8ba9e7e2d0
DTSTAMP:20200602T173219Z
CREATED:20200413T175029Z
DESCRIPTION:Topic: Quantitative decompositions of Lipschitz mappings\n\nSpe
aker: Guy C. David\, Ball State University\n\nGiven a Lipschitz map\, it i
s often useful to chop the domain into pieces on which the map has simple
behavior. For example\, depending on the dimensions of source and target\,
one may ask for pieces on which the map behaves like a bi-Lipschitz embed
ding or like a linear projection. For many issues\, it is even more useful
if this decomposition is quantitative\, i.e.\, with bounds independent of
the particular map or spaces involved. After surveying the question of bi
-Lipschitz decomposition\, we will discuss the more complicated case in wh
ich dimension decreases\, e.g.\, for maps from $\mathbb{R}^3$ to $\mathbb{
R}^2$.\n\nThis is recent joint work with Raanan Schul\, improving a previo
us result of Azzam-Schul.
DTSTART:20200512T150000Z
DTEND:20200512T160000Z
LAST-MODIFIED:20200515T011813Z
LOCATION:https://theias.zoom.us/j/562592856
SUMMARY:Analysis Seminar
URL:https://www.ias.edu/node/110496
END:VEVENT
BEGIN:VEVENT
UID:8bbb8b08-7f82-411f-917e-5df30ad374b4
DTSTAMP:20200602T173219Z
CREATED:20200506T110655Z
DESCRIPTION:Topic: Generative Modeling by Estimating Gradients of the Data
Distribution\n\nSpeaker: Stefano Ermon\, Stanford University\n\nVideo: htt
ps://video.ias.edu/machinelearning/2020/0512-StefanoErmon\n\nExisting gene
rative models are typically based on explicit representations of probabili
ty distributions (e.g.\, autoregressive or VAEs) or implicit sampling proc
edures (e.g.\, GANs). We propose an alternative approach based on modeling
directly the vector field of gradients of the data distribution (scores).
Our framework allows flexible energy-based model architectures\, requires
no sampling during training or the use of adversarial training methods. U
sing annealed Langevin dynamics\, we produces samples comparable to GANs o
n MNIST\, CelebA and CIFAR-10 datasets\, achieving a new state-of-the-art
inception score of 8.91 on CIFAR-10. Finally\, I will discuss challenges i
n evaluating bias and generalization in generative models.
DTSTART:20200512T160000Z
DTEND:20200512T173000Z
LAST-MODIFIED:20200515T011850Z
LOCATION:Remote Access Only - see link below
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/110876
END:VEVENT
BEGIN:VEVENT
UID:9724a822-9786-4f09-8cac-c77275985f26
DTSTAMP:20200602T173219Z
CREATED:20200506T112306Z
DESCRIPTION:Topic: The Simplicity Conjecture\n\nSpeaker: Daniel Cristofaro-
Gardiner\, University of California\, Santa Cruz\; von Neumann Fellow\, Sc
hool of Mathematics\n\nIn the 60s and 70s\, there was a flurry of activity
concerning the question of whether or not various subgroups of homeomorph
ism groups of manifolds are simple\, with beautiful contributions by Fathi
\, Kirby\, Mather\, Thurston\, and many others. A funnily stubborn case th
at remained open was the case of area-preserving homeomorphisms of surface
s. For example\, for balls of dimension at least 3\, the relevant group wa
s shown to be simple by work of Fathi in 1980\; but\, the answer in the tw
o-dimensional case\, asked by Mather in the 70s\, was not known. We recent
ly answered Mather's question by showing that the group of compactly suppo
rted area-preserving homeomorphisms of the two-disc is in fact not simple.
After surveying the history described above\, I will give a very gentle i
ntroduction to some of the key ideas in our proof\; what is crucial is the
fact that the 2-ball with its volume form is a symplectic manifold. Our w
ork underscores that it is natural to study continuous symplectic geometry
\, and I will briefly explain what this means.
DTSTART:20200513T213000Z
DTEND:20200513T230000Z
LAST-MODIFIED:20200509T185444Z
LOCATION:Remote Access Only
SUMMARY:Mathematical Conversations
URL:https://www.ias.edu/node/110881
END:VEVENT
BEGIN:VEVENT
UID:b041e7d4-8f71-4ef8-a93d-6a851067338c
DTSTAMP:20200602T173219Z
CREATED:20200420T122707Z
DESCRIPTION:Topic: A geometric view on Iwasawa theory\n\nSpeaker: Mladen D
imitrov\, Université de Lille\n\nVideo: https://video.ias.edu/puias/2020/0
514-MladenDimitrov\n\nWe will investigate the geometry of the p-adic eigen
curve at classical points where the Galois representation is locally trivi
al at p\, and will give applications to Iwasawa and Hida theories.
DTSTART:20200514T183000Z
DTEND:20200514T193000Z
LAST-MODIFIED:20200515T011934Z
LOCATION:https://theias.zoom.us/j/959183254
SUMMARY:Joint IAS/Princeton University Number Theory Seminar
URL:https://www.ias.edu/node/110596
END:VEVENT
BEGIN:VEVENT
UID:afac2f9f-ea54-40a1-b3f5-2442d62263b3
DTSTAMP:20200602T173219Z
CREATED:20191025T211502Z
DESCRIPTION:Topic: MathZero\, The Classification Problem\, and Set-Theoreti
c Type Theory\n\nSpeaker: David McAllester\, Toyota Technological Institut
e at Chicago \n\nVideo: https://video.ias.edu/machinelearning/2020/0514-Da
vidMcAllester\n\nAlphaZero learns to play go\, chess and shogi at a superh
uman level through self play given only the rules of the game. This raises
the question of whether a similar thing could be done for mathematics ---
a MathZero. MathZero would require a formal foundation and an objective.
We propose the foundation of set-theoretic dependent type theory and an ob
jective defined in terms of the classification problem --- the problem of
classifying concept instances up to isomorphism. Isomorphism is central to
the structure of mathematics. Mathematics is organized around concepts su
ch as graphs\, groups\, topological spaces and manifolds each of which is
associated with a notion of isomorphism. Each concept is associated with a
classification problem --- the problem of enumerating the instances of a
given concept up to isomorphism. The natural numbers arise as the solution
to the classification problem for finite sets. In this talk we attempt to
set the stage for MathZero by giving what we believe to be the first isom
orphism inference rules for set-theoretic dependent type theory with propo
sitional set-theoretic equality. The presentation is intended to be access
ible to mathematicians with no prior exposure to type theory.
DTSTART:20200514T190000Z
DTEND:20200514T203000Z
LAST-MODIFIED:20200515T012016Z
LOCATION:Remote Access Only - see link below
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/106041
END:VEVENT
BEGIN:VEVENT
UID:7307f355-faf7-4bcc-8f9d-7c1b7f842451
DTSTAMP:20200602T173219Z
CREATED:20200420T142606Z
DESCRIPTION:Topic: Reflections on Cylindrical Contact Homology\n\nSpeaker:
Jo Nelson\, Rice University\n\nVideo: https://video.ias.edu/puias/2020/051
5-JoNelson\n\nThis talk beings with a light introduction\, including some
historical anecdotes to motivate the development of this Floer theoretic m
achinery for contact manifolds some 25 years ago. I will discuss joint wor
k with Hutchings which constructs nonequivariant and a family Floer equiva
riant version of contact homology. Both theories are generated by two copi
es of each Reeb orbit over Z and capture interesting torsion information.
I will explain the need for an obstruction bundle gluing correction term i
n the expression of the differential in the presence of contractible Reeb
orbits\, which is essential even in the simple example of an ellipsoid. I
will then explain how one can recover the original cylindrical theory prop
osed by Eliashberg-Givental-Hofer via our constructions.
DTSTART:20200515T130000Z
DTEND:20200515T143000Z
LAST-MODIFIED:20200518T001716Z
LOCATION:https://princeton.zoom.us/j/745635914
SUMMARY:IAS/PU-Montreal-Tel-Aviv Symplectic Geometry Seminar
URL:https://www.ias.edu/node/110621
END:VEVENT
BEGIN:VEVENT
UID:3749bc6c-e20d-470e-8e6f-06b91567070b
DTSTAMP:20200602T173219Z
CREATED:20200508T134428Z
DESCRIPTION:Topic: General Discussion\n\n
DTSTART:20200515T190000Z
DTEND:20200515T200000Z
LAST-MODIFIED:20200508T134428Z
LOCATION:Remote Event
SUMMARY:Princeton University HSC Group Discussion
URL:https://www.ias.edu/node/110911
END:VEVENT
BEGIN:VEVENT
UID:e28d8ef4-2596-473d-ab3d-e60eaa390aff
DTSTAMP:20200602T173219Z
CREATED:20200423T185014Z
DESCRIPTION:Topic: The Non-Stochastic Control Problem\n\nSpeaker: Elad Haza
n\, Princeton University\n\nVideo: https://video.ias.edu/csdm/2020/0518-El
adHazan\n\nLinear dynamical systems are a continuous subclass of reinforce
ment learning models that are widely used in robotics\, finance\, engineer
ing\, and meteorology. Classical control\, since the work of Kalman\, has
focused on dynamics with Gaussian i.i.d. noise\, quadratic loss functions
and\, in terms of provably efficient algorithms\, known systems and observ
ed state. We'll discuss how to apply new machine learning methods which re
lax all of the above: efficient control with adversarial noise\, general l
oss functions\, unknown systems\, and partial observation.\n\nBased on a s
eries of works with Naman Agarwal\, Nataly Brukhim\, Karan Singh\, Sham Ka
kade\, Max Simchowitz\, Cyril Zhang\, Paula Gradu\, John Hallman
DTSTART:20200518T150000Z
DTEND:20200518T160000Z
LAST-MODIFIED:20200518T180314Z
LOCATION:https://theias.zoom.us/j/360043913
SUMMARY:Computer Science/Discrete Mathematics Seminar I
URL:https://www.ias.edu/node/110751
END:VEVENT
BEGIN:VEVENT
UID:535acc0d-4bf8-44e3-8632-87ad4ac1aed9
DTSTAMP:20200602T173219Z
CREATED:20200506T112735Z
DESCRIPTION:Topic: Square function estimate for the cone in R^3\n\nSpeaker:
Hong Wang\, Member\, School of Mathematics\n\nVideo: https://video.ias.ed
u/analysis/2020/05/18-HongWang\n\nWe prove a sharp square function estimat
e for the cone in R^3 and consequently the local smoothing conjecture for
the wave equation in 2+1 dimensions. The proof uses induction on scales an
d an incidence estimate for points and tubes.\n\nThis is joint work with L
arry Guth and Ruixiang Zhang.
DTSTART:20200518T150000Z
DTEND:20200518T160000Z
LAST-MODIFIED:20200518T180249Z
LOCATION:Remote Access via Zoom videoconferencing (link below)
SUMMARY:Analysis Seminar
URL:https://www.ias.edu/node/110886
END:VEVENT
BEGIN:VEVENT
UID:73ed1c17-fdf1-46af-a19f-13c3feb1600b
DTSTAMP:20200602T173219Z
CREATED:20200508T135712Z
DESCRIPTION:Topic: Binary Stars & Exoplanets @ Birmingham\n\nSpeaker: Amaur
y Triaud\, University of Birmingham\n\nI will start by describing our depa
rtment at Birmingham and the type of research we do\, before detailing som
e of the exoplanet research activities of my group. We will focus on two p
articular topics: low-mass eclipsing binaries in order to build reliable m
ass/radius relationships\, including a double-line eclipsing binary brown
dwarf\, the only second ever identified. Following this\, I will show prel
iminary results from our ongoing radial-velocity search for circumbinary p
lanets including multiple exoplanet candidates.
DTSTART:20200518T161500Z
DTEND:20200518T171500Z
LAST-MODIFIED:20200508T135712Z
LOCATION:Virtual Meeting
SUMMARY:Princeton University Exoplanet Discussion Group
URL:https://www.ias.edu/node/110916
END:VEVENT
BEGIN:VEVENT
UID:e42e5446-bbb0-4a8a-9472-2f8637d42277
DTSTAMP:20200602T173219Z
CREATED:20190802T151502Z
DESCRIPTION:Topic: Probing the Tensor Consistency Relation with the CMB Bis
pectrum\n\nSpeaker: Adriaan Duivenvoorden\, Princeton University\n\nIn thi
s talk I will introduce the CMB bispectrum\, discuss recent constraints an
d upcoming observations and\, in particular\, focus on how the bispectrum
can be used to investigate the tensor consistency relation: a relatively u
nexplored type of primordial non-Gaussianity. After a general exposition o
f bispectrum estimation\, I will move on to the tensor consistency relatio
n. I will argue why upcoming CMB data is particularly well-suited to test
the relation: low resolution B-mode data are complementary to high-resolut
ion E-mode and temperature data. Lastly\, I will explain how existing esti
mation techniques have to be adapted for this new type of non-Gaussianity
and discuss progress towards an actual constraint.
DTSTART:20200518T163000Z
DTEND:20200518T180000Z
LAST-MODIFIED:20200514T151917Z
LOCATION:Until further notice\, meetings will be held remotely on Zoom.
SUMMARY:Institute for Advanced Study/Princeton University Early Universe/Co
smology Lunch Discussion - Speaker 1
URL:https://www.ias.edu/node/102651
END:VEVENT
BEGIN:VEVENT
UID:e5efc9e2-7367-4e16-a890-5426b2c2e65f
DTSTAMP:20200602T173219Z
CREATED:20200514T151519Z
DESCRIPTION:Topic: Predicting Electron Gas into Dark Matter-only Simulation
s\n\nSpeaker: Leander Thiele\, Princeton University\n\nForward simulating
the behaviour of baryonic components under gravity as well as astrophysica
l effects is a challenging problem. However\, there is a clear need for ac
curate theoretical predictions of observables sourced by baryonic physics\
, for example the tSZ and kSZ effects. As an alternative to time-intensive
hydrodynamic simulations\, we demonstrate a viable technique to predict s
everal properties of the electron gas into dark matter-only simulations us
ing convolutional neural networks. We provide some new ideas to overcome c
hallenges arising from the sparsity of the data.
DTSTART:20200518T163000Z
DTEND:20200518T180000Z
LAST-MODIFIED:20200514T151519Z
LOCATION:Until further notice\, meetings will be held remotely on Zoom.
SUMMARY:Institute for Advanced Study/Princeton University Early Universe/Co
smology Lunch Discussion - Speaker 2
URL:https://www.ias.edu/node/110966
END:VEVENT
BEGIN:VEVENT
UID:f8f54e29-a610-4580-9196-3d90b32d0f92
DTSTAMP:20200602T173219Z
CREATED:20200515T121054Z
DESCRIPTION:Topic: Neural SDEs: Deep Generative Models in the Diffusion Lim
it\n\nSpeaker: Maxim Raginsky\, University of Illinois Urbana-Champaign\n
\nVideo: https://video.ias.edu/tml/2020/0519-MaximRaginsky\n\nIn deep gene
rative models\, the latent variable is generated by a time-inhomogeneous M
arkov chain\, where at each time step we pass the current state through a
parametric nonlinear map\, such as a feedforward neural net\, and add a sm
all independent Gaussian perturbation. In this talk\, based on joint work
with Belinda Tzen\, I will discuss the diffusion limit of such models\, wh
ere we increase the number of layers while sending the step size and the n
oise variance to zero. I will first provide a unified viewpoint on both sa
mpling and variational inference in such generative models through the len
s of stochastic control. Then I will show how we can quantify the expressi
veness of diffusion-based generative models. Specifically\, I will prove t
hat one can efficiently sample from a wide class of terminal target distri
butions by choosing the drift of the latent diffusion from the class of mu
ltilayer feedforward neural nets\, with the accuracy of sampling measured
by the Kullback-Leibler divergence to the target distribution. Finally\, I
will briefly discuss a scheme for unbiased\, finite-variance simulation i
n such models. This scheme can be implemented as a deep generative model w
ith a random number of layers.
DTSTART:20200519T160000Z
DTEND:20200519T173000Z
LAST-MODIFIED:20200519T183213Z
LOCATION:Remote Access Only - see link below
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/110971
END:VEVENT
BEGIN:VEVENT
UID:175f70c5-5769-43c4-aa9b-24f7cc771d61
DTSTAMP:20200602T173219Z
CREATED:20200511T183237Z
DESCRIPTION:Topic: Conley's fundamental theorem of dynamical systems\n\nSpe
aker: Amie Wilkinson\, University of Chicago\n\nIn 1978\, Charles Conley c
lassified all continuous dynamical systems. His theorem\, dubbed the 'fund
amental theorem of dynamical systems' states that the orbits of any contin
uous map on a compact metric space fall into two classes: gradient-like an
d recurrent. When the recurrent part is factored out\, the dynamics appear
to be gradient-like. While one might wonder how a theorem that applies to
every continuous map could be of any use\, it plays a foundational role i
n many deep results.
DTSTART:20200520T213000Z
DTEND:20200520T230000Z
LAST-MODIFIED:20200518T134724Z
LOCATION:Remote Access Only
SUMMARY:Mathematical Conversations
URL:https://www.ias.edu/node/110951
END:VEVENT
BEGIN:VEVENT
UID:857a3745-07ad-4de2-b139-74ced54bd63b
DTSTAMP:20200602T173219Z
CREATED:20200507T143913Z
DESCRIPTION:Topic: Iwasawa theory and Bloch-Kato conjecture for unitary gro
ups\n\nSpeaker: Xin Wan\, Morningside Center of Mathematics\, Chinese Acad
emy of Sciences\n\nVideo: https://video.ias.edu/jointnumbertheory/2020/052
1-XinWan\n\nWe describe a new method to study Eisenstein family and Iwasaw
a theory on unitary groups over totally real fields of general signatures.
As a consequence we prove that if the central L-value of a cuspidal eigen
form on the unitary group twisted by a CM character is 0\, then the corres
ponding Selmer group has positive rank. The method also has a byproduct th
e p-adic functional equations for p-adic L-functions and p-adic families o
f Eisenstein series on unitary groups.
DTSTART:20200521T130000Z
DTEND:20200521T140000Z
LAST-MODIFIED:20200521T162146Z
LOCATION:https://theias.zoom.us/j/959183254
SUMMARY:Joint IAS/Princeton University Number Theory Seminar
URL:https://www.ias.edu/node/110896
END:VEVENT
BEGIN:VEVENT
UID:6abcd660-366b-4200-b64e-37e4919e9630
DTSTAMP:20200602T173220Z
CREATED:20200506T110159Z
DESCRIPTION:Topic: Forecasting Epidemics and Pandemics\n\nSpeaker: Roni Ros
enfeld\, Carnegie Mellon University\n\nVideo: https://video.ias.edu/tml/20
20/0521-RoniRosenfeld\n\nEpidemiological forecasting is critically needed
for decision making by national and local governments\, public health offi
cials\, healthcare institutions and the general public. The Delphi group a
t Carnegie Mellon University was founded in 2012 to advance the theory and
technological capability of epidemiological forecasting\, and to promote
its role in decision making\, both public and private. Our long term visio
n is to make epidemiological forecasting as useful and universally accepte
d as weather forecasting is today. I will describe some of the methods we
developed over the past eight year for forecasting flu\, dengue and other
epidemics\, and the challenges we faced in adapting these method to the CO
VID pandemic in the past few months.
DTSTART:20200521T190000Z
DTEND:20200521T200000Z
LAST-MODIFIED:20200521T221956Z
LOCATION:Remote Access Only - see link below
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/110871
END:VEVENT
BEGIN:VEVENT
UID:43593691-cf8c-4dcf-a434-b6ac39e3f793
DTSTAMP:20200602T173220Z
CREATED:20200420T142731Z
DESCRIPTION:Topic: Mirrors of curves and their Fukaya categories\n\nSpeaker
: Denis Auroux\, Harvard University\n\nHomological mirror symmetry predict
s that the derived category of coherent sheaves on a curve has a symplecti
c counterpart as the Fukaya category of a mirror space. However\, with the
exception of elliptic curves\, this mirror is usually a symplectic Landau
-Ginzburg model\, i.e. a non-compact manifold equipped with the extra data
of a 'stop' in its boundary at infinity. Most of the talk will focus on a
family of Landau-Ginzburg models which provide mirrors to curves in (C*)^
2 or in toric surfaces (or more generally to hypersurfaces in toric variet
ies)\, and their fiberwise wrapped Fukaya categories (joint work with Moha
mmed Abouzaid). I will then discuss more a speculative way of constructing
mirrors of curves without Landau-Ginzburg models\, involving a new flavor
of Lagrangian Floer theory in trivalent configurations of Riemann surface
s (joint work with Alexander Efimov and Ludmil Katzarkov).
DTSTART:20200522T130000Z
DTEND:20200522T143000Z
LAST-MODIFIED:20200526T182702Z
LOCATION:https://princeton.zoom.us/j/745635914
SUMMARY:IAS/PU-Montreal-Paris-Tel-Aviv Symplectic Geometry Zoominar
URL:https://www.ias.edu/node/110626
END:VEVENT
BEGIN:VEVENT
UID:3007e70f-8d6d-4b22-a707-df3cf24e0a9e
DTSTAMP:20200602T173220Z
CREATED:20200515T212108Z
DESCRIPTION:Topic: General Discussion\n\n
DTSTART:20200522T190000Z
DTEND:20200522T200000Z
LAST-MODIFIED:20200515T212108Z
LOCATION:Remote Event
SUMMARY:Princeton University HSC Group Discussion
URL:https://www.ias.edu/node/110996
END:VEVENT
BEGIN:VEVENT
UID:499bb84c-942b-466b-89c6-aca21d562aa3
DTSTAMP:20200602T173220Z
CREATED:20200513T150837Z
DESCRIPTION:Topic: An application of integers of the 12th cyclotomic field
in the theory of phase transitions\n\nSpeaker: Alik Mazel\, AMC Health\n\n
The construction of pure phases from ground states is performed for $ u >
u_*(d)$ for all values of $d$ except for 39 special ones. For values $d$ w
ith a single equivalence class all periodic ground states generate the cor
responding pure phase which provides a complete description of extreme Gib
bs measures (complete phase diagram). For a general $d$ we prove that at l
east one class of ground states generates pure phases and propose an algor
ithm that decides\, after finitely many iterations\, which classes of grou
nd states generate pure phases. We cojecture that in case of several class
es only one of them generates pure phases which is confirmed by (numerical
) application of our algorithm to several (relatively small) values of $d$
.
DTSTART:20200525T150000Z
DTEND:20200525T160000Z
LAST-MODIFIED:20200518T133844Z
LOCATION:Remote Access via Zoom videoconferencing (link below)
SUMMARY:Analysis Seminar
URL:https://www.ias.edu/node/110961
END:VEVENT
BEGIN:VEVENT
UID:cefb61f2-559f-43b2-8365-6ee5d6ef2db8
DTSTAMP:20200602T173220Z
CREATED:20200518T135028Z
DESCRIPTION:Topic: Emerging symmetries in statistical physics systems\n\nSp
eaker: Hugo Duminil-Copin\, Université de Genève/IHES\n\nA great achieveme
nt of physics in the second half of the twentieth century has been the pre
diction of conformal symmetry of the scaling limit of critical statistical
physics systems. Around the turn of the millenium\, the mathematical unde
rstanding of this fact has progressed tremendously in two dimensions with
the introduction of the Schramm-Loewner Evolution and the proofs of confor
mal invariance of the Ising model and dimers. Nevertheless\, the understan
ding is still restricted to very specific models. In this talk\, we will g
ently introduce the notion of conformal invariance of lattice systems by t
aking the example of percolation models. We will also explain some recent
and partial progress in the direction of proving conformal invariance for
a large class of such models.
DTSTART:20200527T213000Z
DTEND:20200527T230000Z
LAST-MODIFIED:20200526T121254Z
LOCATION:Remote Access Only
SUMMARY:Mathematical Conversations
URL:https://www.ias.edu/node/111001
END:VEVENT
BEGIN:VEVENT
UID:c5f9fa8c-328b-4a94-ba14-83d198fbc8ac
DTSTAMP:20200602T173220Z
CREATED:20200507T144300Z
DESCRIPTION:Topic: Joint equidistribution of adelic torus orbits and famili
es of twisted L-functions\n\nSpeaker: Farrell Brumley\, Université Sorbonn
e Paris Nord\n\nVideo: https://video.ias.edu/puias/2020/0518-FarrellBrumle
y\n\nThe classical Linnik problems are concerned with the equidistribution
of adelic torus orbits on the homogeneous spaces attached to inner forms
of GL2\, as the discriminant of the torus gets large. When specialized\, t
hese problems admit beautiful classical interpretations\, such as the equi
distribution of integer points on spheres\, of Heegner points or packets o
f closed geodesics on the modular surface\, or of supersingular reductions
of CM elliptic curves. In the mid 20th century\, Linnik and his school es
tablished the equidistribution of many of these classical variants through
his ergodic method\, under a congruence condition on the discriminants mo
dulo a fixed auxiliary prime. More recently\, the Waldspurger formula and
subconvex estimates on L-functions were used to remove these congruence co
nditions\, and provide effective power-savings rates.\n\nIn their 2006 ICM
address\, Michel and Venkatesh proposed a variant of this problem in whic
h one considers the product of two distinct inner forms of GL2\, along wit
h a diagonally embedded torus. One can again specialize the setting to obt
ain interesting classical reformulations\, such as the joint equidistribut
ion of integer points on the sphere\, together with the shape of the ortho
gonal lattice. This hybrid context has received a great deal of attention
recently in the dynamics community\, where\, for instance\, the latter pro
blem was solved by Aka\, Einsiedler\, and Shapira\, under supplementary co
ngruence conditions modulo two fixed primes\, using as critical input the
joinings theorem of Einsiedler and Lindenstrauss.\n\nIn joint (ongoing) wo
rk with Valentin Blomer\, we remove the supplementary congruence condition
s in the joint equidistribution problem\, conditionally on the Riemann Hyp
othesis\, while obtaining a logarithmic rate of convergence. The proof use
s Waldsurger’s theorem and estimates of fractional moments of L-functions
in the family of class group twists.
DTSTART:20200528T140000Z
DTEND:20200528T150000Z
LAST-MODIFIED:20200528T183157Z
LOCATION:https://theias.zoom.us/j/959183254
SUMMARY:Joint IAS/Princeton University Number Theory Seminar
URL:https://www.ias.edu/node/110901
END:VEVENT
BEGIN:VEVENT
UID:901d8098-2cf8-41c2-a05e-db0f4224445a
DTSTAMP:20200602T173220Z
CREATED:20200420T142839Z
DESCRIPTION:Topic: Duality for Rabinowitz-Floer homology\n\nSpeaker: Alex O
ancea\, Institut de Mathématiques de Jussieu-Paris Rive Gauche\n\nI will e
xplain a duality theorem with products in Rabinowitz-Floer homology. This
has a bearing on string topology and explains a number of dualities that h
ave been observed in that setting. Joint work in progress with Kai Cielieb
ak and Nancy Hingston.
DTSTART:20200529T130000Z
DTEND:20200529T143000Z
LAST-MODIFIED:20200526T121546Z
LOCATION:https://princeton.zoom.us/j/745635914
SUMMARY:IAS/PU-Montreal-Paris-Tel-Aviv Symplectic Geometry Zoominar
URL:https://www.ias.edu/node/110631
END:VEVENT
BEGIN:VEVENT
UID:0b6a386f-9fd9-425a-a805-ea57f511e585
DTSTAMP:20200602T173220Z
CREATED:20200420T132103Z
DESCRIPTION:Topic: Winding for Wave Maps\n\nSpeaker: Max Engelstein\, Unive
rsity of Minnesota\n\nVideo: https://video.ias.edu/analysis/2020/0601-MaxE
ngelstein\n\nWave maps are harmonic maps from a Lorentzian domain to a Rie
mannian target. Like solutions to many energy critical PDE\, wave maps can
develop singularities where the energy concentrates on arbitrary small sc
ales but the norm stays bounded. Zooming in on these singularities yields
a harmonic map (called a soliton or bubble) in the weak limit. One fundame
ntal question is whether this weak limit is unique\, that is to say\, whet
her different bubbles may appear as the limit of different sequences of re
scalings. We show by example that uniqueness may not hold if the target ma
nifold is not analytic. Our construction is heavily inspired by Peter Topp
ing's analogous example of a ``winding' bubble in harmonic map heat flow.
However\, the Hamiltonian nature of the wave maps will occasionally necess
itate different arguments.\n\nThis is joint work with Dana Mendelson (U Ch
icago).
DTSTART:20200601T150000Z
DTEND:20200601T160000Z
LAST-MODIFIED:20200602T150012Z
LOCATION:Remote Access via Zoom videoconferencing (link below)
SUMMARY:Analysis Seminar
URL:https://www.ias.edu/node/110606
END:VEVENT
BEGIN:VEVENT
UID:67948f1b-cf0a-4672-9aff-7778cb64b35d
DTSTAMP:20200602T173220Z
CREATED:20200529T183732Z
DESCRIPTION:Topic: The Kinematics of Globular Cluster Populations in the E-
Mosaics Simulations and Their Implications for the Assembly History of the
Milky Way\n\nSpeaker: Discussion Leader Sebastian Trujillo-Gomez\, Univer
sität Heidelberg\n\n
DTSTART:20200601T150000Z
DTEND:20200601T160000Z
LAST-MODIFIED:20200529T184457Z
LOCATION:Remote via Zoom
SUMMARY:Princeton University Galread [Galactic/Extragalactic Reading Group]
URL:https://www.ias.edu/node/111111
END:VEVENT
BEGIN:VEVENT
UID:7997257a-8038-4210-8e64-a75ebd415af2
DTSTAMP:20200602T173220Z
CREATED:20200529T182022Z
DESCRIPTION:Topic: Formation of Giant Planet Satellites\n\nSpeaker: Konstan
tin Batygin\, California Institute of Technology\n\nRecent analyses have s
hown that the concluding stages of giant planet formation are accompanied
by the development of large-scale meridional flow of gas inside the planet
ary Hill sphere. This circulation feeds a circumplanetary disk that viscou
sly expels gaseous material back into the parent nebula\, maintaining the
system in a quasi-steady state. Here we investigate the formation of natur
al satellites of Jupiter and Saturn within the framework of this newly out
lined picture. We begin by considering the long-term evolution of solid ma
terial\, and demonstrate that the circumplanetary disk can act as a global
dust trap\, where s ~ 0.1-10 mm grains achieve a hydrodynamical equilibri
um\, facilitated by a balance between radial updraft and aerodynamic drag.
This process leads to a gradual increase in the system's metallicity\, an
d eventually culminates in the gravitational fragmentation of the outer re
gions of the solid sub-disk into R ~ 100 km satellitesimals. Subsequently\
, satellite conglomeration ensues via pairwise collisions\, but is termina
ted when disk-driven orbital migration removes the growing objects from th
e satellitesimal feeding zone. The resulting satellite formation cycle can
repeat multiple times\, until it is brought to an end by photo-evaporatio
n of the parent nebula. Numerical simulations of the envisioned formation
scenario yield satisfactory agreement between our model and the known prop
erties of the Jovian and Saturnian moons.
DTSTART:20200601T161500Z
DTEND:20200601T171500Z
LAST-MODIFIED:20200529T182022Z
LOCATION:Virtual Meeting
SUMMARY:Princeton University Exoplanet Discussion Group
URL:https://www.ias.edu/node/111106
END:VEVENT
BEGIN:VEVENT
UID:05916874-2ead-4feb-9d62-4d9931b36951
DTSTAMP:20200602T173220Z
CREATED:20200529T145923Z
DESCRIPTION:Topic: FlowPM: Distributed TensorFlow Implementation of the Cos
mological N-body Solver\n\nSpeaker: Chirag Modi\, University of California
\, Berkeley\n\n
DTSTART:20200601T163000Z
DTEND:20200601T180000Z
LAST-MODIFIED:20200529T190436Z
LOCATION:Until further notice\, meetings will be held remotely on Zoom.
SUMMARY:Institute for Advanced Study/Princeton University Early Universe/Co
smology Lunch Discussion
URL:https://www.ias.edu/node/111091
END:VEVENT
BEGIN:VEVENT
UID:0d92aa90-f0d4-4d5a-b6a4-71d730649ce2
DTSTAMP:20200602T173220Z
CREATED:20190925T151501Z
DESCRIPTION:
DTSTART:20200602T120000Z
DTEND:20200605T210000Z
LAST-MODIFIED:20200322T195954Z
LOCATION:White-Levy
SUMMARY:CANCELED: NES Conference
URL:https://www.ias.edu/node/105006
END:VEVENT
BEGIN:VEVENT
UID:f8d4e01f-8d00-45b0-81ce-caa4491d8cb8
DTSTAMP:20200602T173220Z
CREATED:20200522T165815Z
DESCRIPTION:Topic: Mathematics formalization for mathematicians\n\nSpeaker:
Patrick Massot\, Université Paris-Sud\n\nA growing number of mathematicia
ns are having fun explaining mathematics to computers using proof assistan
t softwares. This process is called formalization. For instance\, together
with Kevin Buzzard and Johan Commelin\, I recently formalized enough topo
logy and algebra to define Scholze's perfectoid spaces to a computer. In t
his talk I'll describe what formalization looks like\, what kind of things
it teaches us\, and how it could even turn out to be useful (in our usual
sense of 'useful'). This will not be a talk about foundations of mathemat
ics\, and I will stick to elementary examples.
DTSTART:20200603T213000Z
DTEND:20200603T230000Z
LAST-MODIFIED:20200601T125349Z
LOCATION:Remote Access Only
SUMMARY:Mathematical Conversations
URL:https://www.ias.edu/node/111041
END:VEVENT
BEGIN:VEVENT
UID:51cf0570-abfb-4560-886a-b0347d951ddd
DTSTAMP:20200602T173220Z
CREATED:20200519T123912Z
DESCRIPTION:Topic: Digital Transformation of Societal Scale Systems\n\nSpea
ker: Shankar Sastry\, University of California\, Berkeley\n\nOpportunities
abound for the development of new technologies and business models to add
ress some of the most pressing problems of society. Most notably\, the int
egration of IoT\, Data Analytics and Machine Learning and Cloud Computing
into societal scale infrastructures\, such as energy\, transportation\, co
mmunications and financial systems. Of course\, the issues of transforming
societal systems is accompanied by issues of economic models for transfor
mation\, privacy\, (cyber)security and fairness considerations. Indeed\, t
he area of “mechanism design” for societal scale systems is a key feature
in transitioning the newest technologies. Mechanism Design has traditional
ly been in the domain of economics\; however\, the rich interplay of cogni
tive science\, machine learning\, and privacy and cybersecurity makes for
the creation of a new and rich discipline\, which we are calling Digital T
ransformation.\n\nIn this talk\, I will present an intellectual framework
for designing Digital Transformation of Societal Systems\, combining eleme
nts of learning\, game theory\, cybersecurity and utility based privacy em
bodied in a new Institute the C3 Digital Transformation Institute with par
tners Berkeley\, CMU\, Chicago\, UIUC\, MIT\, and Princeton.\n\nSpeaker's
home page: https://people.eecs.berkeley.edu/~sastry
DTSTART:20200604T190000Z
DTEND:20200604T200000Z
LAST-MODIFIED:20200528T130206Z
LOCATION:Remote Access Only - see link below
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/111006
END:VEVENT
BEGIN:VEVENT
UID:7c28d3d9-2bd6-4c77-b987-3e7c9a54a291
DTSTAMP:20200602T173220Z
CREATED:20200508T180330Z
DESCRIPTION:Topic: Dynamical generalizations of the Prime Number Theorem an
d disjointness of additive and multiplicative actions\n\nSpeaker: Florian
Richter\, Northwestern University\n\nOne of the fundamental challenges in
number theory is to understand the intricate way in which the additive and
multiplicative structures in the integers intertwine. We will explore a d
ynamical approach to this topic. After introducing a new dynamical framewo
rk for treating questions in multiplicative number theory\, we will presen
t an ergodic theorem which contains various classical number-theoretic res
ults\, such as the Prime Number Theorem\, as special cases. This naturally
leads to a formulation of an extended form of Sarnak's conjecture\, which
deals with the disjointness of actions of (N\,+) and (N\,*).\n\nThis talk
is based on joint work with Vitaly Bergelson.
DTSTART:20200604T190000Z
DTEND:20200604T200000Z
LAST-MODIFIED:20200519T123508Z
LOCATION:https://theias.zoom.us/j/959183254
SUMMARY:Joint IAS/Princeton University Number Theory Seminar
URL:https://www.ias.edu/node/110931
END:VEVENT
BEGIN:VEVENT
UID:2d61476a-a4ad-4fc8-8496-7ae603d533dd
DTSTAMP:20200602T173220Z
CREATED:20200504T152739Z
DESCRIPTION:Topic: Three Short Research Talks\n\nSpeaker: Morgan Weiler\, J
oé Brendel\, Abror Pirnapasov\n\nMorgan Weiler\, Rice University: Infinite
staircases of symplectic embeddings of ellipsoids into Hirzebruch surface
s\n\nGromov nonsqueezing tells us that symplectic embeddings are governed
by more complex obstructions than volume. In particular\, in 2012\, McDuff
-Schlenk computed the embedding capacity function of the ball\, whose valu
e at a is the size of the smallest four-dimensional ball into which the el
lipsoid E(1\,a) symplectically embeds. They found that it contains an “inf
inite staircase” of piecewise-linear sections accumulating from below to t
he golden ratio to the fourth power. However\, infinite staircases seem to
be rare for more general targets. Work of Cristofaro-Gardiner-Holm-Mandin
i-Pires suggests that\, up to scaling\, there are only finitely many ratio
nal symplectic toric manifolds whose embedding capacity functions contain
infinite staircases\, while Usher has found infinitely many irrational pol
ydisks with infinite staircases. Using ECH capacities in conjunction with
the methods of McDuff-Schlenk\, we will explain how we have found several
infinite families of Hirzebruch surfaces whose embedding capacity function
s we expect to contain an infinite staircase. Many of these staircases are
“descending” rather than “ascending.' This is joint work with Maria Berto
zzi\, Tara Holm\, Emily Maw\, Dusa McDuff\, Grace Mwakyoma\, and Ana Rita
Pires.\n\nJoé Brendel\, University of Neuchâtel : Real Lagrangian Tori in
toric symplectic manifolds\n\nIn this talk we will be addressing the quest
ion whether a given Lagrangian torus in a toric monotone symplectic manifo
ld can be realized as the fixed point set of an anti-symplectic involution
(in which case it is called 'real'). In the case of toric fibres\, the an
swer depends on the geometry of the moment polytope of the ambient manifol
d. In the case of the Chekanov torus\, the answer is always no. This can b
e proved using displacement energy and versal deformations.\n\nAbror Pirna
pasov\, Ruhr-Universität Bochum : Reeb orbits that force topological entro
py\n\nA transverse link in a contact 3-manifold forces topological entropy
if every Reeb flow possessing this link as a set of periodic orbits has p
ositive topological entropy. We will explain how cylindrical contact homol
ogy on the complement of transverse links can be used to show that certain
transverse links force topological entropy. As an application\, we show t
hat on every closed contact 3-manifold exists transverse knots that force
topological entropy. We also generalize to the category of Reeb flows a be
autiful result due to Denvir and Mackay\, which says that if a Riemannian
metric on the two-dimensional torus has a contractible closed geodesic the
n its geodesic flow has positive topological entropy. All this is joint wo
rks with Marcelo R.R. Alves\, Umberto L. Hryniewicz and Pedro A.S. Salomão
DTSTART:20200605T130000Z
DTEND:20200605T143000Z
LAST-MODIFIED:20200529T170833Z
LOCATION:https://princeton.zoom.us/j/745635914
SUMMARY:IAS/PU-Montreal-Paris-Tel-Aviv Symplectic Geometry Zoominar
URL:https://www.ias.edu/node/110831
END:VEVENT
BEGIN:VEVENT
UID:800ba06e-4513-4403-a08a-20e64036c7b1
DTSTAMP:20200602T173220Z
CREATED:20200529T145447Z
DESCRIPTION:Topic: General Discussion\n\n
DTSTART:20200605T190000Z
DTEND:20200605T200000Z
LAST-MODIFIED:20200529T145447Z
LOCATION:Remote Event
SUMMARY:Princeton University HSC Group Discussion
URL:https://www.ias.edu/node/111086
END:VEVENT
BEGIN:VEVENT
UID:5985351b-1a20-4138-8dc8-8b16cf5edf78
DTSTAMP:20200602T173220Z
CREATED:20200601T110927Z
DESCRIPTION:Topic: TBD\n\nSpeaker: Aleksander Mądry\, Massachusetts Institu
te of Technology\n\n
DTSTART:20200609T160000Z
DTEND:20200609T170000Z
LAST-MODIFIED:20200601T110927Z
LOCATION:Remote Access Only - see link below
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/111116
END:VEVENT
BEGIN:VEVENT
UID:3144f77b-6970-4a96-95b7-86a173326418
DTSTAMP:20200602T173220Z
CREATED:20200522T165936Z
DESCRIPTION:Topic: TBA\n\nSpeaker: Michael Hutchings\, University of Califo
rnia - Berkeley\n\n
DTSTART:20200610T213000Z
DTEND:20200610T230000Z
LAST-MODIFIED:20200522T165936Z
LOCATION:Remote Access Only
SUMMARY:Mathematical Conversations
URL:https://www.ias.edu/node/111046
END:VEVENT
BEGIN:VEVENT
UID:7dcfc8a0-6820-45b1-b28b-c30bfcfcec0e
DTSTAMP:20200602T173220Z
CREATED:20200506T113428Z
DESCRIPTION:Topic: TBA\n\nSpeaker: Michael I. Jordan\, University of Califo
rnia\, Berkeley\n\n
DTSTART:20200611T190000Z
DTEND:20200611T203000Z
LAST-MODIFIED:20200521T145423Z
LOCATION:Remote Access Only - see link below
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/110891
END:VEVENT
BEGIN:VEVENT
UID:2ae006e5-caff-4f99-aee7-117e2932c9e2
DTSTAMP:20200602T173220Z
CREATED:20190604T151501Z
DESCRIPTION:
DTSTART:20200612T120000Z
DTEND:20200613T210000Z
LAST-MODIFIED:20200211T211220Z
LOCATION:White-Levy
SUMMARY:NES Colophons Workshop
URL:https://www.ias.edu/node/100506
END:VEVENT
BEGIN:VEVENT
UID:8fb6d137-d95a-457b-905a-72b2e0d77cf0
DTSTAMP:20200602T173220Z
CREATED:20200504T152913Z
DESCRIPTION:Topic: TBA\n\nSpeaker: Mark McLean\, Stony Brook University\n\n
DTSTART:20200612T130000Z
DTEND:20200612T143000Z
LAST-MODIFIED:20200522T131315Z
LOCATION:https://princeton.zoom.us/j/745635914
SUMMARY:IAS/PU-Montreal-Paris-Tel-Aviv Symplectic Geometry Zoominar
URL:https://www.ias.edu/node/110836
END:VEVENT
BEGIN:VEVENT
UID:5329ae04-0d0e-43ea-8818-06df34ea9729
DTSTAMP:20200602T173220Z
CREATED:20200522T170057Z
DESCRIPTION:Topic: TBA\n\nSpeaker: Gigliola Staffilani\, Massachusetts Inst
itutue of Technology\n\n
DTSTART:20200617T213000Z
DTEND:20200617T230000Z
LAST-MODIFIED:20200522T170057Z
LOCATION:Remote Access Only
SUMMARY:Mathematical Conversations
URL:https://www.ias.edu/node/111051
END:VEVENT
BEGIN:VEVENT
UID:ef531267-0479-4f3c-8a43-a2490ab24328
DTSTAMP:20200602T173220Z
CREATED:20200521T145613Z
DESCRIPTION:Topic: TBA\n\nSpeaker: Csaba Szepesvári\, University of Alberta
\n\n
DTSTART:20200618T190000Z
DTEND:20200618T203000Z
LAST-MODIFIED:20200521T145613Z
LOCATION:Remote Access Only - see link below
SUMMARY:Seminar on Theoretical Machine Learning
URL:https://www.ias.edu/node/111031
END:VEVENT
BEGIN:VEVENT
UID:1a7016ac-4e91-404a-894b-2e894db8d897
DTSTAMP:20200602T173220Z
CREATED:20200504T153003Z
DESCRIPTION:Topic: TBA\n\nSpeaker: Igor Uljarevic\, University of Belgrade
\n\n
DTSTART:20200619T130000Z
DTEND:20200619T143000Z
LAST-MODIFIED:20200522T131340Z
LOCATION:https://princeton.zoom.us/j/745635914
SUMMARY:IAS/PU-Montreal-Paris-Tel-Aviv Symplectic Geometry Zoominar
URL:https://www.ias.edu/node/110841
END:VEVENT
BEGIN:VEVENT
UID:12f3b1cc-b0f4-4243-9d3a-667d944690b5
DTSTAMP:20200602T173220Z
CREATED:20200522T170347Z
DESCRIPTION:Topic: TBA\n\nSpeaker: Lisa Piccirillo\, Brandeis/Massachusetts
Institutue of Technology\n\n
DTSTART:20200624T213000Z
DTEND:20200624T230000Z
LAST-MODIFIED:20200522T170347Z
LOCATION:Remote Access Only
SUMMARY:Mathematical Conversations
URL:https://www.ias.edu/node/111056
END:VEVENT
BEGIN:VEVENT
UID:e0b66ebb-2582-4f25-b52e-db80b5313c8a
DTSTAMP:20200602T173220Z
CREATED:20200504T153057Z
DESCRIPTION:Topic: TBA\n\nSpeaker: Ailsa Keating\, Cambridge University\n\n
DTSTART:20200626T130000Z
DTEND:20200626T143000Z
LAST-MODIFIED:20200529T152511Z
LOCATION:https://princeton.zoom.us/j/745635914
SUMMARY:IAS/PU-Montreal-Paris-Tel-Aviv Symplectic Geometry Zoominar
URL:https://www.ias.edu/node/110846
END:VEVENT
BEGIN:VEVENT
UID:0884cff0-729a-47fb-9ea7-4e477fbd15de
DTSTAMP:20200602T173220Z
CREATED:20200522T172515Z
DESCRIPTION:Topic: TBA\n\nSpeaker: Albert Fathi\, Georgia Institute of Tech
nology\n\n
DTSTART:20200701T213000Z
DTEND:20200701T230000Z
LAST-MODIFIED:20200522T172515Z
LOCATION:Remote Access Only
SUMMARY:Mathematical Conversations
URL:https://www.ias.edu/node/111061
END:VEVENT
BEGIN:VEVENT
UID:923b0bec-8521-4e87-a54e-552f5441ad46
DTSTAMP:20200602T173220Z
CREATED:20180808T185302Z
DESCRIPTION:
DTSTART:20200901T161500Z
DTEND:20200901T180000Z
LAST-MODIFIED:20180808T185302Z
LOCATION:Dilworth Room
SUMMARY:Institute for Advanced Study / Princeton University Bahcall Lunch
URL:https://www.ias.edu/node/89086
END:VEVENT
BEGIN:VEVENT
UID:0f27991e-4c3e-4c55-a7f2-d85ebb01ffd9
DTSTAMP:20200602T173220Z
CREATED:20180808T185024Z
DESCRIPTION:
DTSTART:20200908T161500Z
DTEND:20200908T180000Z
LAST-MODIFIED:20180808T185024Z
LOCATION:Dilworth Room
SUMMARY:Institute for Advanced Study / Princeton University Bahcall Lunch
URL:https://www.ias.edu/node/89011
END:VEVENT
BEGIN:VEVENT
UID:44a0495c-c192-426b-87b3-a87372dc6f2a
DTSTAMP:20200602T173220Z
CREATED:20180808T185024Z
DESCRIPTION:
DTSTART:20200915T161500Z
DTEND:20200915T180000Z
LAST-MODIFIED:20180808T185024Z
LOCATION:Dilworth Room
SUMMARY:Institute for Advanced Study / Princeton University Bahcall Lunch
URL:https://www.ias.edu/node/89016
END:VEVENT
BEGIN:VEVENT
UID:577f7e80-cc27-4151-8999-ab4474160813
DTSTAMP:20200602T173220Z
CREATED:20200110T165145Z
DESCRIPTION:
DTSTART:20200921T130000Z
DTEND:20200921T130000Z
LAST-MODIFIED:20200110T165145Z
LOCATION:
SUMMARY:IAS School of Mathematics Term I Begins
URL:https://www.ias.edu/node/107721
END:VEVENT
BEGIN:VEVENT
UID:55e2ce73-2a07-4d1a-821e-8da79268e17e
DTSTAMP:20200602T173220Z
CREATED:20180808T185024Z
DESCRIPTION:
DTSTART:20200922T161500Z
DTEND:20200922T180000Z
LAST-MODIFIED:20180808T185024Z
LOCATION:Dilworth Room
SUMMARY:Institute for Advanced Study / Princeton University Bahcall Lunch
URL:https://www.ias.edu/node/89021
END:VEVENT
BEGIN:VEVENT
UID:7fd6943e-1104-4880-97b4-9375d4f19d7f
DTSTAMP:20200602T173220Z
CREATED:20180808T185024Z
DESCRIPTION:
DTSTART:20200929T161500Z
DTEND:20200929T180000Z
LAST-MODIFIED:20180808T185024Z
LOCATION:Dilworth Room
SUMMARY:Institute for Advanced Study / Princeton University Bahcall Lunch
URL:https://www.ias.edu/node/89026
END:VEVENT
BEGIN:VEVENT
UID:91e51e9c-929a-442b-99f9-a09f2ffb0259
DTSTAMP:20200602T173220Z
CREATED:20180808T185024Z
DESCRIPTION:
DTSTART:20201006T161500Z
DTEND:20201006T180000Z
LAST-MODIFIED:20180808T185024Z
LOCATION:Dilworth Room
SUMMARY:Institute for Advanced Study / Princeton University Bahcall Lunch
URL:https://www.ias.edu/node/89031
END:VEVENT
BEGIN:VEVENT
UID:1e2c0ab9-f9d9-4ac7-899a-adc2ae2a269f
DTSTAMP:20200602T173220Z
CREATED:20180808T185024Z
DESCRIPTION:
DTSTART:20201013T161500Z
DTEND:20201013T180000Z
LAST-MODIFIED:20180808T185024Z
LOCATION:Dilworth Room
SUMMARY:Institute for Advanced Study / Princeton University Bahcall Lunch
URL:https://www.ias.edu/node/89036
END:VEVENT
BEGIN:VEVENT
UID:c0b42cba-1bf6-41e2-84fb-ed60a70d16a1
DTSTAMP:20200602T173220Z
CREATED:20180808T185024Z
DESCRIPTION:
DTSTART:20201020T161500Z
DTEND:20201020T180000Z
LAST-MODIFIED:20180808T185024Z
LOCATION:Dilworth Room
SUMMARY:Institute for Advanced Study / Princeton University Bahcall Lunch
URL:https://www.ias.edu/node/89041
END:VEVENT
BEGIN:VEVENT
UID:dd36e863-7546-41a1-bf0d-be5c49caa1ca
DTSTAMP:20200602T173220Z
CREATED:20180808T185024Z
DESCRIPTION:
DTSTART:20201027T161500Z
DTEND:20201027T180000Z
LAST-MODIFIED:20180808T185024Z
LOCATION:Dilworth Room
SUMMARY:Institute for Advanced Study / Princeton University Bahcall Lunch
URL:https://www.ias.edu/node/89046
END:VEVENT
BEGIN:VEVENT
UID:35903a4b-5e59-473f-a718-bd529a5781fc
DTSTAMP:20200602T173220Z
CREATED:20180808T185024Z
DESCRIPTION:
DTSTART:20201103T171500Z
DTEND:20201103T190000Z
LAST-MODIFIED:20180808T185024Z
LOCATION:Dilworth Room
SUMMARY:Institute for Advanced Study / Princeton University Bahcall Lunch
URL:https://www.ias.edu/node/89051
END:VEVENT
BEGIN:VEVENT
UID:778f907f-f778-4a64-9ef3-a914c0dd57ae
DTSTAMP:20200602T173220Z
CREATED:20180808T185024Z
DESCRIPTION:
DTSTART:20201110T171500Z
DTEND:20201110T190000Z
LAST-MODIFIED:20180808T185024Z
LOCATION:Dilworth Room
SUMMARY:Institute for Advanced Study / Princeton University Bahcall Lunch
URL:https://www.ias.edu/node/89056
END:VEVENT
BEGIN:VEVENT
UID:41fe7927-7aa2-4d7c-94a2-0856ca7faaab
DTSTAMP:20200602T173220Z
CREATED:20180808T185024Z
DESCRIPTION:
DTSTART:20201117T171500Z
DTEND:20201117T190000Z
LAST-MODIFIED:20180808T185024Z
LOCATION:Dilworth Room
SUMMARY:Institute for Advanced Study / Princeton University Bahcall Lunch
URL:https://www.ias.edu/node/89061
END:VEVENT
BEGIN:VEVENT
UID:e6b671c0-0f52-4cde-9ef4-cd1b8a5ee2b8
DTSTAMP:20200602T173220Z
CREATED:20180808T185024Z
DESCRIPTION:
DTSTART:20201124T171500Z
DTEND:20201124T190000Z
LAST-MODIFIED:20180808T185024Z
LOCATION:Dilworth Room
SUMMARY:Institute for Advanced Study / Princeton University Bahcall Lunch
URL:https://www.ias.edu/node/89066
END:VEVENT
BEGIN:VEVENT
UID:751f1015-f1db-4d4d-9ebf-c34a3298dba0
DTSTAMP:20200602T173220Z
CREATED:20180808T185024Z
DESCRIPTION:
DTSTART:20201201T171500Z
DTEND:20201201T190000Z
LAST-MODIFIED:20180808T185024Z
LOCATION:Dilworth Room
SUMMARY:Institute for Advanced Study / Princeton University Bahcall Lunch
URL:https://www.ias.edu/node/89071
END:VEVENT
BEGIN:VEVENT
UID:4117ab65-1291-45d2-9b7a-b04abe9fc5c2
DTSTAMP:20200602T173220Z
CREATED:20180808T185024Z
DESCRIPTION:
DTSTART:20201208T171500Z
DTEND:20201208T190000Z
LAST-MODIFIED:20180808T185024Z
LOCATION:Dilworth Room
SUMMARY:Institute for Advanced Study / Princeton University Bahcall Lunch
URL:https://www.ias.edu/node/89076
END:VEVENT
BEGIN:VEVENT
UID:3077cf5b-4104-4117-ba59-8b39b955e6b3
DTSTAMP:20200602T173220Z
CREATED:20180808T185024Z
DESCRIPTION:
DTSTART:20201215T171500Z
DTEND:20201215T190000Z
LAST-MODIFIED:20180808T185024Z
LOCATION:Dilworth Room
SUMMARY:Institute for Advanced Study / Princeton University Bahcall Lunch
URL:https://www.ias.edu/node/89081
END:VEVENT
BEGIN:VEVENT
UID:494d5ff6-d3e6-4798-ab52-0d99bf5f66da
DTSTAMP:20200602T173220Z
CREATED:20200110T165522Z
DESCRIPTION:
DTSTART:20201218T230000Z
DTEND:20201218T230000Z
LAST-MODIFIED:20200110T165522Z
LOCATION:
SUMMARY:IAS School of Mathematics Term I Ends
URL:https://www.ias.edu/node/107731
END:VEVENT
BEGIN:VEVENT
UID:ba356055-e4d8-4f99-aa2c-d09abe9aaa9b
DTSTAMP:20200602T173220Z
CREATED:20200110T165737Z
DESCRIPTION:
DTSTART:20210111T140000Z
DTEND:20210111T140000Z
LAST-MODIFIED:20200110T165737Z
LOCATION:
SUMMARY:IAS School of Mathematics Term II Begins
URL:https://www.ias.edu/node/107736
END:VEVENT
BEGIN:VEVENT
UID:6d2eec99-f720-49a0-99df-9425b855570d
DTSTAMP:20200602T173220Z
CREATED:20200110T165817Z
DESCRIPTION:
DTSTART:20210409T220000Z
DTEND:20210409T220000Z
LAST-MODIFIED:20200110T165817Z
LOCATION:
SUMMARY:IAS School of Mathematics Term II Ends
URL:https://www.ias.edu/node/107741
END:VEVENT
END:VCALENDAR