Machine Learning Summer School 2006 - Taipei

Graphical Models, Variational Methods, and Message-Passing

author: Martin J. Wainwright, University of California in Berkeley

Categories

Top: Computer Science: Machine Learning: Graphical Models

You might be experiencing some problems with Your Video player.

*Slides*
0:05	Graphical models, message-passing algorithms, and variational methods: Part I
1:06	Introduction
2:26	Some broad questions
5:35	Outline
6:16	Example: Hidden Markov models
9:53	Example: Image processing and computer vision (I)
14:03	Example: Computer vision (II)
14:27	Stereo pairs: Dom St. Stephan, Passau
17:15	Example: Graphical codes for communication
20:31	Graphical codes and decoding (continued)
20:39	Outline
21:21	Undirected graphical models (Markov random fields)
23:27	Factorization and Markov properties
28:23	Illustrative example: Ising/Potts model for images
29:52	Directed graphical models
32:11	Illustrative example: Hidden Markov model (HMM)
33:02	Factor graph representations
43:43	Representational equivalence: Factorization and Markov property
46:33	Outline
47:52	Computational problems of interest
52:17	Elimination algorithm (I)
54:42	Elimination algorithm (II;A)
55:21	Elimination algorithm (II;B)
55:48	Elimination algorithm (III)
55:52	Elimination algorithm (IV)
56:00	Elimination algorithm (V)
59:15	Summary of elimination algorithm
62:52	Sum-product algorithm on (junction) trees
69:02	Sum-product on trees (I)
72:46	Sum-product on trees (II)
73:03	Sum-product on trees (III)
73:11	Summary: sum-product algorithm on a tree
74:48	Max-product algorithm on trees (I)

Lecture rating

People found this lecture:
Worth seeing
because it is:
Valuable and informative
Well presented
Easily understandable
Acceptably recorded
You need to login to cast your vote.

Report a problem or upload files

If you have found a problem with this lecture or would like to send us extra material, articles, exercises, etc., please use our ticket system to describe your request and upload the data.
Enter your e-mail into the 'Cc' field, and we will keep you updated with your request's status.