Efficient Algorithm for Binary Quadratic Problem by Column Generation and Quantum Annealing

TL;DR

This paper introduces a novel algorithm combining column generation and quantum annealing to efficiently approximate solutions for NP-hard binary quadratic problems, significantly reducing computational time compared to existing methods.

Contribution

The paper presents a new hybrid approach that integrates quantum annealing with column generation to improve solution efficiency for binary quadratic problems.

Findings

Achieved 2.7 to 1000 times faster solutions than existing solvers.

Successfully applied quantum annealing to reduce bottlenecks in column generation.

Demonstrated effectiveness on benchmark binary quadratic problems.

Abstract

We propose an efficient algorithm that combines column generation and quantum annealing to solve binary quadratic problems. Binary quadratic problems are difficult to solve because they are NP-hard. An attempt to solve binary quadratic problems efficiently by column generation has been studied, but it demands successively solving quadratic unconstrained binary optimization problems. We solve the bottleneck by using quantum annealing or simulated annealing. Our results demonstrate a good approximate solution obtained in 2.7 to 1000 times shorter computational time to use column generation and quantum annealing to solve binary quadratic problems than the existing fast solver.