Optimization and benchmarking of the thermal cycling algorithm

TL;DR

This paper benchmarks and enhances the thermal cycling algorithm for complex optimization problems, showing it rivals advanced methods like parallel tempering and surpasses simpler heuristics such as simulated annealing.

Contribution

The paper provides a comprehensive parameter tuning and improvement of the thermal cycling algorithm, demonstrating its competitive performance in nonconvex optimization.

Findings

Thermal cycling algorithm closely competes with parallel tempering.

Outperforms simulated annealing significantly.

Parameter tuning enhances algorithm efficiency.

Abstract

Optimization plays a significant role in many areas of science and technology. Most of the industrial optimization problems have inordinately complex structures that render finding their global minima a daunting task. Therefore, designing heuristics that can efficiently solve such problems is of utmost importance. In this paper we benchmark and improve the thermal cycling algorithm [Phys. Rev. Lett. 79, 4297 (1997)] that is designed to overcome energy barriers in nonconvex optimization problems by temperature cycling of a pool of candidate solutions. We perform a comprehensive parameter tuning of the algorithm and demonstrate that it competes closely with other state-of-the-art algorithms such as parallel tempering with isoenergetic cluster moves, while overwhelmingly outperforming more simplistic heuristics such as simulated annealing.