STA 294 Lecture Notes

STA 294 Lecture Notes

1/13/99: Introduction

1/15/99: Graph Preliminaries

1/20/99: Undirected Graphical Models (1/27/99: added new sections on minimality and perfect clique sequences.)

1/22/99: A Proof for Intersection (1/27/99: "separation" changed to "intersection".)

1/27/99: Directed Graphical Models (1/27/99: gamma changed to x_i)

2/5/99-2/8/99: Exact Inference Algorithms (2/8/99: Talk: Learning from what you don't observe in STA 295)

2/10/99-2/15/99: Exact Inference Lecture Slides

2/17/99: Assessing Feasibility and Structure in Belief Networks

2/19/99: Probability Assessment in Belief Networks

2/22/99: Belief network applications: Similarity Networks, Value of Information and Pathfinder.

2/24/99: Belief network applications: Answer Wizard and Office Assistant.

2/26/99: Recap of assessment techniques, Hugin conflict measure and overview of applications.

3/1/99: Forward Sampling, Likelihood Weighting, Bounded Variance, Gibbs sampling.

3/3/99: Convergence in Gibbs Sampling, Boyen and Koller.

3/5/99: No class.

3/8/99: Approximation in Junction Trees, Variational Approximation in QMR-DT.

3/10/99: No class.

3/12/99: Dynamic Discretization

3/16/99 - 3/20/99: Spring Break

3/22/99: Conditional gaussian distributions, introduction to learning.

Relevant paper by Kevin Murphy: Filtering, Smoothing and the Junction Tree Algorithm.

3/24/99: Parameter estimation with complete data.

The WinBUGS file for the thumbtack example: beta-bugs.odc

3/26/99: Parameter estimation with incomplete data.

3/29/99: Bayesian prediction with incomplete data, Structure scoring metrics for complete data.

3/31/99: Structure optimization with complete data.

D. Heckerman. A tutorial on learning with Bayesian networks. Technical Report MSR-TR-95-06, Microsoft Research, March, 1995 (in course reader).

Moises Goldszmidt and Nir Friedman.
Tutorial slides from AAAI-98' - "Learning Bayesian Networks from Data"

4/2/99: Structure optimization with incomplete data.

Scoring:
D. Chickering and D. Heckerman. Efficient Approximations for the Marginal Likelihood of Bayesian Networks With Hidden Variables Technical Report MSR-TR-96-08

EM:
Ghahramani, Z. (1995)
Factorial Learning and the EM Algorithm
In G. Tesauro, D.S. Touretzky, and J. Alspector (eds.), Advances in Neural Information Processing Systems 7

4/5/99: Classifiers

Relevant papers:
Friedman, Geiger + Goldszmidt: Bayesian Network Classifiers

4/7/99: Clustering and Learning local structure

Relevant papers:
Friedman + Goldszmidt: Learning Bayesian Networks with Local Structure.

D. Chickering, D. Heckerman, C. Meek A Bayesian Approach to Learning Bayesian Networks with Local Structure Technical Report MSR-TR-97-07

4/9/99: The normal Wishart distribution: Learning gaussian belief networks.

Relevant papers/books:
Morris DeGroot: Optimal Statistical Decisions, Sections 5.4-5.6 and 9.9-9.13.

D. Heckerman, D. Geiger, D. Chickering. Learning Bayesian networks: The Combination of Knowledge and Statistical Data. Technical Report MSR-TR-94-09.

B. Thiesson, C. Meek, D. Chickering, D. Heckerman Learning Mixtures of DAG Models Technical Report MSR-TR-97-30

D. Geiger and D. Heckerman. Parameter Priors for Directed Acyclic Graphical Models and the Characterization of Several Probability Distributions Technical Report MSR-TR-98-67

4/12/99: Learning Dynamic Belief Networks.

Relevant papers
Nir Friedman, Kevin Murphy, and Stuart Russell: Learning the Structure of Dynamic Probabilistic Networks
UAI '98.
Roweis, S.T. and Ghahramani, Z. (1997)
A Unifying Review of Linear Gaussian Models

Ghahramani, Z. (1998)
Learning Dynamic Bayesian Networks
In C.L. Giles and M. Gori (eds.), Adaptive Processing of Sequences and Data Structures . Lecture Notes in Artificial Intelligence, 168-197. Berlin: Springer-Verlag.