Lazy Explanation-Based Approximation for Probabilistic Logic Programming

TL;DR

This paper presents a lazy, explanation-based approximation method for probabilistic logic programming that efficiently computes probability bounds, outperforming existing approaches in speed and accuracy.

Contribution

It introduces a novel lazy approximation technique that uses only key parts of the program for faster, anytime inference with reliable probability bounds.

Findings

Outperforms state-of-the-art approximate inference methods

Provides fast, anytime probability bounds

Demonstrates effectiveness through experimental evaluation

Abstract

We introduce a lazy approach to the explanation-based approximation of probabilistic logic programs. It uses only the most significant part of the program when searching for explanations. The result is a fast and anytime approximate inference algorithm which returns hard lower and upper bounds on the exact probability. We experimentally show that this method outperforms state-of-the-art approximate inference.