BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//MIT Center for Statistics - ECPv4.5.9//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:MIT Center for Statistics
X-ORIGINAL-URL:https://stat.mit.edu
X-WR-CALDESC:Events for MIT Center for Statistics
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20170908T110000
DTEND;TZID=America/New_York:20170908T110000
DTSTAMP:20170824T103055
CREATED:20170801T015203Z
LAST-MODIFIED:20170809T154616Z
UID:1676-1504868400-1504868400@stat.mit.edu
SUMMARY:New provable techniques for learning and inference in probabilistic graphical models Andrej Risteski (MIT)
DESCRIPTION:A common theme in machine learning is succinct modeling of distributions over large domains. Probabilistic graphical models are one of the most expressive frameworks for doing this. The two major tasks involving graphical models are learning and inference. Learning is the task of calculating the “best fit” model parameters from raw data\, while inference is the task of answering probabilistic queries for a model with known parameters (e.g. what is the marginal distribution of a subset of variables\, after conditioning on the values of some other variables). Learning can be thought of as finding a graphical model that “explains” the raw data\, while the inference queries extract the “knowledge” the graphical model contains. \nI will focus on a few vignettes from my research which give new provable techniques for these tasks:\n– In the context of learning\, I will talk about method-of-moments techniques for learning noisy-or Bayesian networks\, which are used for modeling the causal structure of diseases and symptoms.\n– In the context of inference\, I will talk about a new understanding of a class of algorithms for calculating partition functions\, called variational methods\, through the\nlens of convex programming hierarchies. Time permitting\, I will also speak about MCMC methods for sampling from highly multimodal distributions using simulated tempering. \nThe talk will assume no background\, and is meant as a “meet and greet” talk surveying various questions I’ve worked on and am interested in. \n
URL:https://stat.mit.edu/calendar/stochastics-and-statistics-seminar-andrej-risteski/
CATEGORIES:Stochastics and Statistics Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20170914T160000
DTEND;TZID=America/New_York:20170914T170000
DTSTAMP:20170824T103055
CREATED:20170802T135644Z
LAST-MODIFIED:20170804T191404Z
UID:1773-1505404800-1505408400@stat.mit.edu
SUMMARY:Special Stochastics and Statistics Seminar John Cunningham (Columbia University)
DESCRIPTION:
URL:https://stat.mit.edu/calendar/stochastics-statistics-seminar-john-cunningham-columbia/
CATEGORIES:Stochastics and Statistics Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20170915T110000
DTEND;TZID=America/New_York:20170915T110000
DTSTAMP:20170824T103055
CREATED:20170801T015203Z
LAST-MODIFIED:20170816T171754Z
UID:1677-1505473200-1505473200@stat.mit.edu
SUMMARY:Sample complexity of population recovery Yury Polyanskiy (MIT)
DESCRIPTION:In this talk we will first consider a general question of estimating linear functional of the distribution based on the noisy samples from it. We discover that the (two-point) LeCam lower bound is in fact achievable by optimizing bias-variance tradeoff of an empirical-mean type of estimator. \nNext\, we apply this general framework to the specific problem of population recovery. Namely\, consider a random poll of sample size n conducted on a population of individuals\, where each pollee is asked to answer d binary questions. We consider one of the two polling impediments: \n\nin lossy population recovery\, a pollee may skip each question with probability ε;\nin noisy population recovery\, a pollee may lie on each question with probability ε.\n\nGiven n lossy or noisy samples\, the goal is to estimate the probabilities of all 2d binary vectors simultaneously within accuracy δ with high probability. We settle the sample complexity for both models and discover curious phase-transition in the lossy model at ε = 1/2. Our results hinge on complex-analytic methods\, such as Hadamard’s three-lines theorem. \n \nBiography: Yury Polyanskiy is an Associate Professor of Electrical Engineering and Computer Science and a member of LIDS at MIT. Yury received M.S. degree in applied mathematics and physics from the Moscow Institute of Physics and Technology\, Moscow\, Russia in 2005 and Ph.D. degree in electrical engineering from Princeton University\, Princeton\, NJ in 2010. Currently\, his research focuses on basic questions in information theory\, error-correcting codes\, wireless communication and fault tolerant and defect-tolerant circuits. Dr. Polyanskiy won the 2013 NSF CAREER award and 2011 IEEE Information Theory Society Paper Award. \n
URL:https://stat.mit.edu/calendar/stochastics-and-statistics-seminar-yuri-polyanskiy/
CATEGORIES:Stochastics and Statistics Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20170922T110000
DTEND;TZID=America/New_York:20170922T110000
DTSTAMP:20170824T103055
CREATED:20170801T015204Z
LAST-MODIFIED:20170822T180033Z
UID:1678-1506078000-1506078000@stat.mit.edu
SUMMARY:Stochastics and Statistics Seminar Amir Dembo (Stanford University)
DESCRIPTION:
URL:https://stat.mit.edu/calendar/stochastics-and-statistics-seminar-amir-dembo/
CATEGORIES:Stochastics and Statistics Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20170929T110000
DTEND;TZID=America/New_York:20170929T110000
DTSTAMP:20170824T103055
CREATED:20170801T015204Z
LAST-MODIFIED:20170803T200310Z
UID:1679-1506682800-1506682800@stat.mit.edu
SUMMARY:Optimal lower bounds for universal relation\, and for samplers and finding duplicates in streams Jelani Nelson (Harvard University)
DESCRIPTION: Consider the following problem: we monitor a sequence of edgeinsertions and deletions in a graph on n vertices\, so there are N = (n choose 2) possible edges (e.g. monitoring a stream of friend accepts/removals on Facebook). At any point someone may say “query()”\, at which point must output a random edge that exists in the graph at that time from a distribution that is statistically close to uniform. More specifically\, with probability p our edge should come from a distribution close to uniform\, and with probability 1-p our sampler can do anything (i.e. it can say “Fail”\, or output a non-existent edge). The primary goal is to design an algorithm that uses very little memory. Whereas the trivial solution maintains N bits\, keeping track of which edges currently exist\, we show that the optimal space complexity for this problem is Theta(min{N\, log(1/p) * log^2(2 + N / log(1/p))}) bits. The upper bound follows by a very minor modification\nof previous work\, whereas the lower bound is our main novel contribution. \nThis is joint work with Michael Kapralov (EPFL)\, Jakub Pachocki (OpenAI)\, Zhengyu Wang (Harvard)\, David P. Woodruff (IBM Almaden)\, and Mobin Yahyazadeh (Sharif). \n
URL:https://stat.mit.edu/calendar/optimal-lower-bounds-for-universal-relation-and-for-samplers-and-finding-duplicates-in-streams/
CATEGORIES:Stochastics and Statistics Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20171002T093000
DTEND;TZID=America/New_York:20171002T173000
DTSTAMP:20170824T103055
CREATED:20170814T165852Z
LAST-MODIFIED:20170814T165940Z
UID:1867-1506936600-1506965400@stat.mit.edu
SUMMARY:2017 Charles River Lectures on Probability and Related Topics
DESCRIPTION:The Charles River Lectures on Probability and Related Topics will be hosted by Harvard University on Monday\, October 2\, 2017 in Cambridge\, MA. The lectures are jointly organized by Harvard University\, Massachusetts Institute of Technology and Microsoft Research New England for the benefit of the greater Boston area mathematics community. The event features five lectures by distinguished researchers in the areas of probability and related topics. This year’s lectures will be delivered by: \nPaul Bourgade (Courant Institute\, NYU)\nMassimiliano Gubinelli (University of Bonn)\nAndrea Montanari (Stanford University)\nRoman Vershynin (University of California\, Irvine)\nOfer Zeitouni (Weizmann Institute) \nFor questions regarding the event\, please email agontar@math.harvard.edu \n \n
URL:https://stat.mit.edu/calendar/2017-charles-river-lectures-probability-related-topics/
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20171006T110000
DTEND;TZID=America/New_York:20171006T110000
DTSTAMP:20170824T103055
CREATED:20170801T015204Z
LAST-MODIFIED:20170804T192727Z
UID:1680-1507287600-1507287600@stat.mit.edu
SUMMARY:Stochastics and Statistics Seminar Youseff Marzouk (MIT)
DESCRIPTION:
URL:https://stat.mit.edu/calendar/stochastics-and-statistics-seminar-youseff-marzouk/
CATEGORIES:Stochastics and Statistics Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20171013T110000
DTEND;TZID=America/New_York:20171013T110000
DTSTAMP:20170824T103055
CREATED:20170801T015204Z
LAST-MODIFIED:20170804T201216Z
UID:1681-1507892400-1507892400@stat.mit.edu
SUMMARY:Stochastics and Statistics Seminar Galen Reeves (Duke University)
DESCRIPTION:
URL:https://stat.mit.edu/calendar/stochastics-and-statistics-seminar-galen-reeves/
CATEGORIES:Stochastics and Statistics Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20171020T110000
DTEND;TZID=America/New_York:20171020T110000
DTSTAMP:20170824T103055
CREATED:20170801T015204Z
LAST-MODIFIED:20170818T141113Z
UID:1682-1508497200-1508497200@stat.mit.edu
SUMMARY:Inference in dynamical systems and the geometry of learning group actions Sayan Mukherjee (Duke)
DESCRIPTION:We examine consistency of the Gibbs posterior for dynamical systems using a classical idea in dynamical systems called the thermodynamic formalism in tracking dynamical systems. We state a variation formulation under which there is a unique posterior distribution of parameters as well as hidden states using using classic ideas from dynamical systems such as pressure and joinings. We use an example of consistency of hidden Markov with infinite lags as an application of our theory. \nWe develop a geometric framework that characterizes the synchronization problem — the problem of consistently registering or aligning a collection of objects. The theory we formulate characterizes the cohomological nature of synchronization based on the classical theory of fibre bundles and that synchronization can be characterized by trivial holonomy. We then develop a twisted cohomology theory to quantify obstructions to synchronization\, this is a discrete version of the twisted cohomology in differential geometry. \nMotivated by our geometric framework\, we study the problem of learning group actions — partitioning a collection of objects based on the local synchronizability of pairwise correspondence relations. A dual interpretation is to learn finitely generated subgroups of an ambient transformation group from noisy observed group elements. A synchronization-based algorithm is also provided\, and we demonstrate its efficacy in a problem in geometric morphometrics\, clustering the molars of primates according to their eating habits. \n \n Biography : Sayan Mukherjee is Professor of Statistical Science\, Mathematics\, Computer Science\, and Biostatistics \& Bioinformatics at Duke University. He received a PhD in 2001 from the Center for Biological and Computational Learning at MI. He was a Sloan-DOE Postdoctoral Fellow in Computational Biology 2001-2004 at the Broad Institute of MIT and Harvard. My research areas cover Bayesian methodology; computational and statistical methods in statistical genetics\, quantitative genetics\, cancer biology\, and morphology; discrete Hodge theory\, geometry and topology in statistical inference; inference in dynamical systems; machine learning; and stochastic topology. biology being cited over 11\,000 times since 2004. \n
URL:https://stat.mit.edu/calendar/stochastics-and-statistic-seminar-sayan-mukherjee/
CATEGORIES:Stochastics and Statistics Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20171027T110000
DTEND;TZID=America/New_York:20171027T110000
DTSTAMP:20170824T103055
CREATED:20170801T015205Z
LAST-MODIFIED:20170804T203057Z
UID:1683-1509102000-1509102000@stat.mit.edu
SUMMARY:Stochastics and Statistic Seminar Amit Daniely (Google)
DESCRIPTION:
URL:https://stat.mit.edu/calendar/stochastics-and-statistic-seminar-amit-daniely/
CATEGORIES:Stochastics and Statistics Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20171103T110000
DTEND;TZID=America/New_York:20171103T110000
DTSTAMP:20170824T103055
CREATED:20170801T015205Z
LAST-MODIFIED:20170804T203159Z
UID:1684-1509706800-1509706800@stat.mit.edu
SUMMARY:Stochastics and Statistic Seminar Joan Bruna Estrach (NYU)
DESCRIPTION:
URL:https://stat.mit.edu/calendar/stochastics-and-statistic-seminar-joan-bruna-estrach/
CATEGORIES:Stochastics and Statistics Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20171117T110000
DTEND;TZID=America/New_York:20171117T110000
DTSTAMP:20170824T103055
CREATED:20170801T015205Z
LAST-MODIFIED:20170807T125213Z
UID:1685-1510916400-1510916400@stat.mit.edu
SUMMARY:Stochastics and Statistic Seminar Alex Dimakis (University of Texas at Austin)
DESCRIPTION:
URL:https://stat.mit.edu/calendar/stochastics-and-statistic-seminar-alex-dimakis/
CATEGORIES:Stochastics and Statistics Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20171201T110000
DTEND;TZID=America/New_York:20171201T110000
DTSTAMP:20170824T103055
CREATED:20170801T015205Z
LAST-MODIFIED:20170807T163242Z
UID:1686-1512126000-1512126000@stat.mit.edu
SUMMARY:Stochastics and Statistic Seminar Susan Murphy (University of Michigan)
DESCRIPTION:
URL:https://stat.mit.edu/calendar/stochastics-and-statistic-seminar-susan-murphy/
CATEGORIES:Stochastics and Statistics Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20171208T110000
DTEND;TZID=America/New_York:20171208T110000
DTSTAMP:20170824T103055
CREATED:20170801T015205Z
LAST-MODIFIED:20170807T162937Z
UID:1687-1512730800-1512730800@stat.mit.edu
SUMMARY:Stochastics and Statistics Seminar Alex Bloemendal (Broad Institute)
DESCRIPTION:
URL:https://stat.mit.edu/calendar/stochastics-and-statistics-seminar-alex-bloemendal/
CATEGORIES:Stochastics and Statistics Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180406T080000
DTEND;TZID=America/New_York:20180406T170000
DTSTAMP:20170824T103055
CREATED:20170803T135349Z
LAST-MODIFIED:20170803T135922Z
UID:1786-1523001600-1523034000@stat.mit.edu
SUMMARY:SDSCon 2018: Statistics and Data Science Center Conference
DESCRIPTION:Coming soon… \n
URL:https://stat.mit.edu/calendar/sdscon-statistics-data-science-center-conference/
LOCATION:75 Amherst Street\, Cambridge\, MA\, 02139\, United States
GEO:42.3603959;-71.0872332
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=75 Amherst Street Cambridge MA 02139 United States;X-APPLE-RADIUS=500;X-TITLE=75 Amherst Street:geo:-71.0872332,42.3603959
CATEGORIES:SDSC Special Events
END:VEVENT
END:VCALENDAR