Probabilistic Horn Abduction and Bayesian Networks

This paper presents a simple framework for Horn clause abduc tion with probabilities associated with hypotheses The framework incorporates assumptions about the rule base and independence as sumptions amongst hypotheses It is shown how any probabilistic knowledge representable in a discrete Bayesian belief network can be represented in this framework The main contribution is in nding a relationship between logical and probabilistic notions of evidential reasoning This provides a useful representation language in its own right providing a compromise between heuristic and epistemic ad equacy It also shows how Bayesian networks can be extended be yond a propositional language This paper also shows how a language with only unconditionally independent hypotheses can represent any probabilistic knowledge and argues that it is better to invent new hy potheses to explain dependence rather than having to worry about dependence in the language Scholar Canadian Institute for Advanced Research Probabilistic Horn abduction and Bayesian networks