Chapter 7

TAXONOMIC HIERARCHIES, CONTINUOUS VARIABLES, AND UNCERTAIN PROBABILITIES

Publisher Summary

This chapter presents propagation rules that, like those developed for causal networks, allow evidence to be pooled through local computations, under local control. The chapter presents an encoding scheme that stores and manipulates the belief distributions and the support messages rather than storing them explicitly as in the case of discrete variables. The encoding scheme is based on three assumptions: (1) all interactions between variables are linear, (2) the sources of uncertainty are normally distributed and are uncorrelated, and (3) the causal network is singly connected. Even though the mathematical derivations rely on these three ...

Get Probabilistic Reasoning in Intelligent Systems now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.