Comparing the hardness of MAX 2-SAT problem instances for quantum and classical algorithms

TL;DR

This study compares the difficulty of MAX 2-SAT problem instances for quantum and classical algorithms, revealing weak correlations in hardness and advocating for a portfolio approach to improve problem-solving efficiency.

Contribution

It provides a numerical analysis of instance hardness for quantum and classical algorithms, highlighting the limitations of small-sized benchmarks and the benefits of a portfolio approach.

Findings

Weak correlation in instance hardness across algorithms

Hardness distribution widens with increasing problem size

Portfolio approach can mitigate extremely hard instances

Abstract

An algorithm for a particular problem may find some instances of the problem easier and others harder to solve, even for a fixed input size. We numerically analyse the relative hardness of MAX 2-SAT problem instances for various continuous-time quantum algorithms and a comparable classical algorithm. This has two motivations: to investigate whether small-sized problem instances, which are commonly used in numerical simulations of quantum algorithms for benchmarking purposes, are a good representation of larger instances in terms of their hardness to solve, and to determine the applicability of continuous-time quantum algorithms in a portfolio approach, where we take advantage of the variation in the hardness of instances between different algorithms by running them in parallel. We find that, while there are correlations in instance hardness between all of the algorithms considered, they appear weak enough that a portfolio approach would likely be desirable in practice. Our results also show a widening range of hardness of randomly generated instances as the problem size is increased, which demonstrates both the difference in the distribution of hardness at small sizes and the value of a portfolio approach that can reduce the number of extremely hard instances. We identify specific weaknesses of these quantum algorithms that can be overcome with a portfolio approach, such their inability to efficiently solve satisfiable instances (which is easy classically).