Intelligent Probabilistic Inference

Abstract

Publisher Summary This chapter discusses the intelligent probabilistic inference. The analysis of practical probabilistic models on the computer demands a convenient representation for the available knowledge and an efficient algorithm to perform inference. The chapter focuses on arbitrary conditional probability distributions from a given probabilistic model. With given qualitative information about the dependence of the random variables in the model, for a specific conditional expression, quantitative information needed to determine the desired conditional probability distribution can be specified precisely. In this way, the conditional independence present in the model can be exploited to avoid having to construct or manipulate the full joint distribution. These results are extended to include maximal processing when the information available is incomplete and optimal decision making in an uncertain environment. One main application of this research is in the construction of a decision system, an automated tool to assist a decision maker. The influence diagrams processed by the algorithm can be constructed and interpreted by programs within the system. The ability to determine the information needed to answer a given question is critical in such an environment.