View
215
Download
1
Embed Size (px)
Citation preview
CSI 661 - Uncertainty in A.I. Lecture 9 1
CSI 661 - Uncertainty in A.I. Lecture 9 2
CSI 661 - Uncertainty in A.I. Lecture 9 3
Three Main Approaches To Approximate Inference
• MCMC
• Variational Methods
• Loopy belief propagation
CSI 661 - Uncertainty in A.I. Lecture 9 4
Decision Making With BNets
CSI 661 - Uncertainty in A.I. Lecture 9 5
Minimize Risk
CSI 661 - Uncertainty in A.I. Lecture 9 6
Influence Diagrams
CSI 661 - Uncertainty in A.I. Lecture 9 7
Putting a Price on Information
CSI 661 - Uncertainty in A.I. Lecture 9 8
Improvement With Extra Link
CSI 661 - Uncertainty in A.I. Lecture 9 9
Dynamic Belief Networks
CSI 661 - Uncertainty in A.I. Lecture 9 10
DBNet
CSI 661 - Uncertainty in A.I. Lecture 9 11
HMM Algorithms Applied to DBNET
CSI 661 - Uncertainty in A.I. Lecture 9 12
Next Lecture
• Major criticism of BNets as models of human behavior
• Learning Bayes Nets– Parameter estimation– Learning structure