Belief Updating and Learning in Semi-Qualitative Probabilistic Networks
arXiv:1207.1367
Abstract
This paper explores semi-qualitative probabilistic networks (SQPNs) that combine numeric and qualitative information. We first show that exact inferences with SQPNs are NPPP-Complete. We then show that existing qualitative relations in SQPNs (plus probabilistic logic and imprecise assessments) can be dealt effectively through multilinear programming. We then discuss learning: we consider a maximum likelihood method that generates point estimates given a SQPN and empirical data, and we describe a Bayesian-minded method that employs the Imprecise Dirichlet Model to generate set-valued estimates.
Appears in Proceedings of the Twenty-First Conference on Uncertainty in Artificial Intelligence (UAI2005)
References in corpus (8)
- Complexity Results and Approximation Strategies for MAP Explanations
- Elicitation of Probabilities for Belief Networks: Combining Qualitative and Quantitative Information
- From Qualitative to Quantitative Probabilistic Networks
- Exploiting Qualitative Knowledge in the Learning of Conditional Probabilities of Bayesian Networks
- Inference in Polytrees with Sets of Probabilities
- Computing Probability Intervals Under Independency Constraints
- Enhancing QPNs for Trade-off Resolution
- Upgrading Ambiguous Signs in QPNs