On Bayesian Network Approximation by Edge Deletion

TL;DR

This paper introduces a new evidence-based edge deletion method for Bayesian networks that simplifies models for probabilistic inference, providing bounds on approximation quality and demonstrating empirical effectiveness.

Contribution

A novel edge deletion approach for Bayesian networks that leverages evidence to improve approximation quality and offers theoretical bounds on divergence.

Findings

Bounds on KL-divergence highlight evidence impact

Edge deletion improves inference efficiency

Empirical results show promising approximation quality

Abstract

We consider the problem of deleting edges from a Bayesian network for the purpose of simplifying models in probabilistic inference. In particular, we propose a new method for deleting network edges, which is based on the evidence at hand. We provide some interesting bounds on the KL-divergence between original and approximate networks, which highlight the impact of given evidence on the quality of approximation and shed some light on good and bad candidates for edge deletion. We finally demonstrate empirically the promise of the proposed edge deletion technique as a basis for approximate inference.