Course Description

This is a doctoral student seminar covering current topics in applied probability. For the offering of Spring 2006, the topics covered will be at the interface of statistical physics (replica symmetry breaking and spin glasses), probability (local weak convergence and correlation decay), artificial intelligence (belief propagation algorithms), computer science (random K-SAT problem, coloring, average case complexity) and electrical engineering (Low density parity check (LDPC) codes).


The course requires knowledge of optimization, probability and algorithms. A basic course on probability (6.041/6.431) is necessary. An advanced graduate course in probability (6.436J/15.085J) will be very useful.

Course Policy

The course will meet once every week for two hours where students will do presentations. There will not be any exams or homework. The grade of the student will be based on the presentations and class participation.