|
This quarter we will focus on the approximate inference for the graphical model and two specific models, hidden Markov model and Kalman filters.
Homepage for previous quarters
To subscribe to the mailing list, go to the cse590mv mail list
The address of the mailing list is cse590mv AT cs.washington.edu
Thursdays
An Introduction to Graphical Models, Michael I. Jordan and Christopher M. Bishop, preprint
Hybrid Bayesian Networks for Reasoning about Complex Systems, Uri Lerner, Ph.D. thesis, http://robotics.stanford.edu/~uri/Papers/thesis.ps.gz
Week |
Topic |
Paper/Chapter |
Leader |
1 |
Discuss the plan for the quarter |
|
-- |
2 |
Hidden Markov Model (HMM) |
Chap. 12 of Jordon’s book |
|
3 |
Talk by Drew Bagnell at Intel |
|
|
4 |
EM algorithm for HMM |
Chap. 10 of Jordon’s book |
|
5 |
Gaussians for Graphical Model |
Chap. 3 of Lerner’s thesis (optional) Chap. 13 of Jordon’s book |
|
6 |
Gaussians for Graphical Model (cont’d) |
|
|
7 |
Kalman Filtering and Smoothing |
Chap. 15 of Jordon’s book |
|
8 |
Variational method |
Chap. 22 of Jordon’s book |
|
9 |
Variational method (cont'd) |
||
10 |
Variational method (cont'd) |
|