CSE515 Statistical Methods in Computer Science – Spring 2011

Lectures (tentative)

Week Date Topic Reading Slides HW
1 3/28 Introduction, Bayesian network representation 2.1-3, 3.1 (PDF) (Annotated PDF)
3/30 Bayesian network representation cont. 3.1-3 (PDF) (Annotated PDF)
2 4/4 Local probability models 5 (PDF) HW1 out
4/6 Undirected graphical models I 4 (PDF) (Annotated PDF)
3 4/11 Undirected graphical models II + CRFs and applications 4 (PDF) (Annotated PDF)
4/13 Inference: Exact inference 9.1-3 (PDF) (Annotated PDF)
4 4/18 Exact inference 9.4-6 (PDF) (Annotated PDF) HW2 out; HW1 due
4/20 Exact inference: Clique Trees 10.1-4 (PDF) (Annotated PDF)
5 4/25 Learning: parameter estimation 17 (PDF) (Annotated PDF)
4/27 Parameter learning in BNs 17 (PDF) (Annotated PDF)
6 5/2 Structure learning in BNs 18 (PDF) (Annotated PDF) HW3 out; HW2 due
5/4 Partially observed data (learning with missing data) 19 (PDF) (Annotated PDF)
7 5/9 EM algorithm and Applications (PDF) (Annotated PDF)
5/11 Approximate inference: particle based 12 (PDF) (Annotated PDF)
8 5/16 Particle-based approximate inference 12 (PDF) (Annotated PDF) HW3 due
5/18 Global approximate inference 11 (PDF) (Annotated PDF) HW4 out
9 5/23 Learning Undirected Models 20 (PDF) (Annotated PDF)
5/25 Markov Decision Processes (Instructor: Mausam) (PDF)
10 5/30 (memorial day)
6/1 Temporal models (DBNs, HMMs) (PDF) HW4 due