The Graduate University for Advanced Studies (SOKENDAI), Department of Informatics
Second Semester 2007
Probabilistic Models in Informatics
16:30-18:00, Tuesday
The focus of this course is probabilistic models, which play an important role for the modeling of real-world data in various fields of informatics. This course deals with foundations and applications of probability theory, and discusses basic ideas of probabilistic models, such as latent variable models and Bayesian Networks.

Course Plan

1 10/16 Introduction to Probability
2 10/23 Probability Theory and Bayes Theorem
3 10/30 Probability Distributions and Expectation
4 11/06 Likelihood and Bayesian Estimation
5 11/13 Maximum Likelihood
11/20 Canceled
6 11/27 Latent Variable Models
12/04 Canceled
7 12/11 Hidden Markov Models
8 12/18 Review of Related Models
9 01/08 Graphical Models (1)
10 01/15 Graphical Models (2)
11 01/22 Sampling and Markov Chain Monte Carlo
12 01/29 Model Selection
02/05 Canceled for entrance exam
13 02/12 Presentations and Summary

Paper Reading Assignment

1 12/18 Goto Jun Applying Conditional Random Fields to Japanese Morphological Analysis
2 01/08 Hutchatai Chanlekha Nymble: a High-Performance Learning Name-finder
3 01/22 Ikeda Naruki Bayesian Analysis of Empirical Software Engineering Cost Models

Presentation of Experiments