G 8325:
Entropy and Information in Probability

Columbia University
Spring 2009


Handouts and homeworks

Links

Schedule of end-of-semester presentations

TUESDAY, MAY 5, 2-5pm

 

Johannes Ruf                            The central limit theorem in relative entropy

Qinghua Li                                Entropy and free probability

Tomoyuki Ichiba                       Extreme value distributions and relative entropy

Petr Novotny                            Universal portfolios

Subhankar Sadhukhan  ss         Large deviations for AR processes

Timothy Teravainen                  Freidlin-Wentzel large deviations theory

Li Song                                    Entropy in time series models

Chun Yip Yau                          Weak consistency of the MDL principle

 

WEDNESDAY, MAY 6, 2-5pm

 

Emilio Seijo                              Concentration of measure via the entropy method

Ivor Cribben                             Bootstrap and maximum entropy distributions

Henry Lam                               Rare event simulation

George Fellouris                       Distributed hypothesis testing

Greg Wayne                             Information-theoretic ideas in control theory

Kamiar Rahnama                      Mutual information expansions

Yashar Ahmadian                     Relative entropy-like metrics for spike train models

 

Announcement: There will be no class on Tuesday 3 and Tuesday 10 of March

Basic Course Information

G 8325 is a topics course offered by the Statistics Department.
CLASS TIMES: Tues/Thur 2:40-3:55 p.m.
LOCATION: room 1025 SSW bldg.

 

Instructor: Ioannis Kontoyiannis

Email: ik2241 at columbia.edu

Office hours: Tursdays 4-6 p.m., or by arrangement

 

Description:

 

Course will contain a subset of the following:

 

Entropy and information: typical strings and the "asymptotic equipartition property"; entropy as the fundamental compression limit; relative entropy as the optimal error exponent in hypothesis testing; Fisher information as the derivative of the entropy; maximum entropy distributions; basic inequalities

 

Probability: The method of types; the strong law of large numbers via the entropy, the central limit theorem as a version of the second law of thermodynamics, large deviations, Sanov's theorem, high-dimensional projections and statistical mechanics; convergence of Markov chains

 

Special topics: Ergodicity, recurrence properties; the Shannon-McMillan-Breiman theorem; Poisson approximation bounds in terms of relative entropy; information in sigma-algebras and the Hewitt-Savage 0-1 law; entropy and the distribution of prime numbers.

Reference texts

Material will be drawn from various places in the literature, including the books:
    Elements of Information Theory by Cover and Thomas
    Information Theory by Csiszar and Korner, and
    Information theory and Statistics by Kullback

Course requirements/exams

There will be homework assignments every 2-3 weeks. Instead of a final exam, students will have the option of either:

    Giving an oral presentation in class; or

    Doing a project and writing a project report.

A list of possible topics for presentations and projects will be provided by the instructor. Possible projects will cover the whole range from applied computational projects to purely theoretical questions in probability. New research topics will also be introduced along the way.

Prerequisites

Knowledge of basic probability and random processes. No previous knowledge of information theory will be required.


 


Last modified: May 4, 2009