Impact of Graph Structures for QAOA on MaxCut

TL;DR

This study evaluates how graph structure influences the performance of QAOA on MaxCut for small graphs, revealing key structural predictors and providing a benchmark database for future research.

Contribution

It systematically analyzes the impact of graph features on QAOA performance across all small connected graphs, and shares a benchmark dataset.

Findings

Odd-cycles and symmetry predict QAOA success

Performance varies significantly with graph structure

Benchmark database for QAOA experiments

Abstract

The quantum approximate optimization algorithm (QAOA) is a promising method of solving combinatorial optimization problems using quantum computing. QAOA on the MaxCut problem has been studied extensively on specific families of graphs, however, little is known about the algorithm on arbitrary graphs. We evaluate the performance of QAOA at depths at most three on the MaxCut problem for all connected non-isomorphic graphs with at most eight vertices and analyze how graph structure affects QAOA performance. Some of the strongest predictors of QAOA success are the existence of odd-cycles and the amount of symmetry in the graph. The data generated from these studies are shared in a publicly-accessible database to serve as a benchmark for QAOA calculations and experiments. Knowing the relationship between structure and performance can allow us to identify classes of combinatorial problems that are likely to exhibit a quantum advantage.