Beam Search: Faster and Monotonic

TL;DR

This paper introduces a monotonic variant of beam search that guarantees non-increasing solution costs with increasing beam width and demonstrates how distance-to-go estimates enhance solution quality and speed in non-uniform cost domains.

Contribution

It presents a new monotonic beam search algorithm and shows how to incorporate distance-to-go estimates for improved performance.

Findings

Monotonic beam search guarantees non-increasing solution costs.

Distance-to-go estimates improve solution quality and speed.

Practical effectiveness of beam search is enhanced.

Abstract

Beam search is a popular satisficing approach to heuristic search problems that allows one to trade increased computation time for lower solution cost by increasing the beam width parameter. We make two contributions to the study of beam search. First, we show how to make beam search monotonic; that is, we provide a new variant that guarantees non-increasing solution cost as the beam width is increased. This makes setting the beam parameter much easier. Second, we show how using distance-to-go estimates can allow beam search to find better solutions more quickly in domains with non-uniform costs. Together, these results improve the practical effectiveness of beam search.