# You can also click the upper-left icon to select videos from the playlist.
source: Centre International de Rencontres Mathématiques 2016年2月16日
RESEARCH SCHOOL - CIRM MARSEILLE LUMINY
Tutorial Week - Nexus of Information and Computation Theories
January 25 - 29, 2016
Over the past few decades, the theories of computation and information have been studied in separate academic circles, which have worked in near isolation from one another. More recently, several researchers have uncovered close connections between these two disciplines, both in terms of problem formulations and proof techniques. As a result, there is a growing interest in building bridges between these two communities in order to make progress on topics of common interest. The aim of this winter school is to make a concerted effort to broaden and deepen the connections between information theory (IT) and the theory of computation (ToC).
The program will consist of several tutorials, each taught by a leading researcher, with the goal of introducing the key questions, mathematical tools, and open problems in an area. The school will also serve as a prelude to a 2016 Institut Henri Poincaré thematic program of the same title.
Ankur Moitra : Algorithmic Aspects of Inference 53:08
Abstract: Parametric inference is one of the cornerstones of statistics, but much of the classic theory revolves around asymptotic notions of convergence and relies on estimators that are hard to compute (particularly in high-dimensional problems).
In this tutorial, we will explore the following questions:
(1) For some of the fundamental problems in statistics, are there surrogates for the maximum likelihood estimator that also converge at an inverse polynomial rate to the true parameters, but in contrast can be computed efficiently?
(2) Can we establish tradeoffs between sample complexity and computational complexity? And what types of hardness assumptions allow us to explore this space?
We will cover topics such as the method of moments, learning mixture models, tensor decomposition, sparse PCA and matrix/tensor completion.
Recording during the thematic meeting: «Nexus of Information and Computation Theories » theJanuary 25, 2016 at the Centre International de Rencontres Mathématiques (Marseille, France)
Filmmaker: Guillaume Hennenfent
Ankur Moitra : Tensor Decompositions and their Applications 57:54
Ankur Moitra : A Polynomial Time Algorithm for Lossy Population Recovery 54:02
Ankur Moitra : Linear Inverse Problems 1:01:25
Sudeep Kamath : Concentration of Measure - 2 1:06:02
Sudeep Kamath : Concentration of Measure - 1 1:00:28
Sudeep Kamath : Concentration of Measure - 3 53:50
Sudeep Kamath : Concentration of Measure - 4 1:02:46
Anup Rao : Communication Complexity and Information Complexity - 1 1:08:38
Anup Rao : Communication Complexity and Information Complexity - 2 1:08:15
Anup Rao : Communication Complexity and Information Complexity - 3 54:59
Anup Rao : Communication Complexity and Information Complexity - 4 1:00:52
Guy Rothblum : Privacy and Security via Randomized Methods - 1 1:02:23
Guy Rothblum : Privacy and Security via Randomized Methods - 2 1:10:52
Guy Rothblum : Privacy and Security via Randomized Methods - 3 1:04:51
Guy Rothblum : Privacy and Security via Randomized Methods - 4 1:04:42
No comments:
Post a Comment