Difference between revisions of "10-601 GM1"
From Cohen Courses
Jump to navigationJump to search| Line 4: | Line 4: | ||
=== Slides === | === Slides === | ||
| − | * Ziv's lecture: [http://www.cs.cmu.edu/~zivbj/classF14/ | + | * Ziv's lecture: [http://www.cs.cmu.edu/~zivbj/classF14/BNs.pdf Slides in pdf]. |
=== Readings === | === Readings === | ||
Revision as of 13:56, 31 October 2014
This a lecture used in the Syllabus for Machine Learning 10-601 in Fall 2014
Slides
- Ziv's lecture: Slides in pdf.
Readings
- Chap 8.1 and 8.2.2 (Bishop)
- Graphical Models by Michael I. Jordan
Taking home message
- factorization theorem of BN
- Full, independent and intermediate conditional probability models
- Markov blanket
- Learning a BN
- Inference in BN is NP hard
- Approximate inference in BN
