A New Algorithm for Finding MAP Assignments to Belief Networks

TL;DR

This paper introduces a novel algorithm for computing MAP assignments in belief networks by converting them into boolean networks and applying best-first search, with efficiency improvements for poly trees.

Contribution

The paper proposes a new algorithm that simplifies belief networks into boolean networks and efficiently finds MAP assignments, especially in poly trees.

Findings

Algorithm is exponential in general case

Algorithm runs linearly on poly trees

Effective for belief networks with specific structures

Abstract

We present a new algorithm for finding maximum a-posterior) (MAP) assignments of values to belief networks. The belief network is compiled into a network consisting only of nodes with boolean (i.e. only 0 or 1) conditional probabilities. The MAP assignment is then found using a best-first search on the resulting network. We argue that, as one would anticipate, the algorithm is exponential for the general case, but only linear in the size of the network for poly trees.