6.437 -- Inference and Information
Course Description: Introduction to principles of Bayesian and non-Bayesian statistical inference. Hypothesis testing and parameter estimation, sufficient statistics; exponential families. Log-loss inference criterion, entropy and model capacity. Kullback-Leibler distance and information geometry. Asymptotic analysis and large deviations theory. Model order estimation; nonparametric statistics. Computational issues and approximation techniques; Monte Carlo methods. Selected special topics such as universal prediction and compression.

This class is at the Graduate level
Instructor: P. Golland, A. S. Willsky, G. W. Wornell
Prerequisites: 6.041/6.431 or 6.436J

Students's Comments on the Class

From a PhD student in the ORC
I would definitely recommend this course to other students. The course covered a variety of topics in statistical inference and information theory with a good balance between theory and application. The course is intended to provide a strong background for follow-on coursework in Machine Learning and includes some topics from the field. Compared to STATS 220 at Harvard, this course avoided some of the purely theoretical statistical concepts in favor of including more topics from information theory. I think this course is better than the 15.076 and 15.077 combination.

Back to Classes