NCU AI Course

Recent site activity

Chapter 14. Probabilistic Reasoning

Slides-1, Slides-2.

14.1. Representing Knowledge in an Uncertain Domain: Introduction to Bayesian Network
14.2. The Semantics of Bayesian Networks
       Representing the full joint distribution
              A method for constructing Bayesian networks
              Compactness and node ordering
       Conditional independence relations in Bayesian networks: Markov Blanket
14.3. Efficient Representation of Conditional Distributions
              Bayesian nets with continuous variables
Continuous variables with discrete and continuous parents
Discrete variables with continuous parents: probit distribution or logit distribution
14.4. Exact Inference in Bayesian Networks
       Inference by enumeration
       The variable elimination algorithm: avoid repeat computation
       The complexity of exact inference: singly connected or mulltiply connectede
       Clustering algorithms
14.5. Approximate Inference in Bayesian Networks
       Direct sampling methods
              Rejection sampling in Bayesian networks
              Likelihood weighting
       Inference by Markov chain simulation
              The MCMC algorithm
              Why MCMC works