An axiomatic framework for propagating uncertainty in directed acyclic networks (Q2366555)

The authors adapt and generalize the propagation model for probabilistic information of \textit{J. Pearl} [Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference, Morgan and Kaufmann: San Matteo (1988; Zbl 0746.68089)] in such a way that it may also cope with beliefs, conditional information, or possibilities. They start from ideas of \textit{G. Shaefer} and \textit{P. P. Shenoy} [in: \textit{R. T. Schachter} et al. (eds.) Uncertainty in Artificial Intelligence 4, 169-198 Amsterdam (1990)], but add new axioms, consider acyclic graphs instead of hypertrees, and concentrate on operations with valuations.