Bayesian Networks and Causal Modelling .ppt

Bayesian Networks and Causal Modelling Ann Nicholson Overview Introduction to Bayesian Networks (BNs) Summary of BN research projects Varieties of Causal intervention PRICAI2004: K. Korb, L. Hope, A. Nicholson, K. Axnick Learning Causal Structure CaMML software Probability theory for representing uncertainty Assigns a numerical degree of belief between 0 and 1 to facts e.g. “it will rain today” is T/F. P(“it will rain today”) = 0.2 prior probability (unconditional) Posterior probability (conditional) P(“it wil rain today” | “rain is forecast”) = 0.8 Bayes’ Rule: P(H|E) = P(E|H) x P(H) P(E) Bayesian networks A Bayesian Network (BN) represents a probability distribution graphically (directed acyclic graphs) Nodes: random variables, R: “it is raining”, discrete values T/F T: temperature, cts or discrete variable C: colour, discrete values {red,blue,green} Arcs indicate conditional dependencies between variables Bayesian networks (cont.) There is a conditional probability distribution (CPD or CPT) associated with each node. probability of each state given parent states BN inference Evidence: observation of specific state Task: compute the posterior probabilities for query node(s) given evidence. Causal Networks Arcs follow the direction of causal process Causal Networks are always BNs Bayesian Networks arent always causal Early BN-related projects DBNS for discrete monitoring (PhD, 1992) Approximate BN inference algorithms based on a mutual information measure for relevance (with Nathalie Jitnah, 1996-1999) Plan recognition: DBNs for predicting users actions and goals in an adventure game (with David Albrecht, Ingrid Zukerman, 1997-2000) DBNs for ambulation monitoring and fall diagnosis (with biomedical engineering, 1996-2000) Bayesian Poker (with Kevin Korb, 1996-2003) Knowledge Engineering with BNs Seabreeze prediction: joint project with Bureau of Meteorology Comparison of existing simple rule, expert eli