A Caltech Library Service

Optimization and benchmarking of the thermal cycling algorithm

Barzegar, Amin and Kankani, Anuj and Mandrà, Salvatore and Katzgraber, Helmut G. (2021) Optimization and benchmarking of the thermal cycling algorithm. Physical Review E, 104 (3). Art. No. 035302. ISSN 2470-0045. doi:10.1103/physreve.104.035302.

[img] PDF - Published Version
See Usage Policy.

[img] PDF - Accepted Version
See Usage Policy.


Use this Persistent URL to link to this item:


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.

Item Type:Article
Related URLs:
URLURL TypeDescription Paper
Barzegar, Amin0000-0001-6346-8047
Kankani, Anuj0000-0002-7422-9137
Mandrà, Salvatore0000-0002-3908-2694
Katzgraber, Helmut G.0000-0003-3341-9943
Additional Information:© 2021 American Physical Society. (Received 13 January 2021; revised 9 August 2021; accepted 26 August 2021; published 8 September 2021) We thank A. Möbius for useful discussions regarding various aspects of the thermal cycling algorithm, and also providing code for the dsrm. The authors also acknowledge Jonathan Machta for critically reviewing the manuscript. H.G.K. thanks Dr. Pimple Popper for visualizations of energy landscapes. We thank Texas A&M University, NASA Ames Research Center, and Microsoft Quantum Group for providing access to computational resources. This work is supported in part by the Office of the Director of National Intelligence (ODNI), Intelligence Advanced Research Projects Activity (IARPA), via MIT Lincoln Laboratory Air Force Contract No. FA8721-05-C-0002. S.M. also acknowledges the support from the Intelligence Advanced Research Projects Activity (IARPA) (IARPA IAA 1198) and the Defense Advanced Research Projects Agency (DARPA) (IAA 8839, Annex 125). The views and conclusions contained herein are those of the authors and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of ODNI, IARPA, DARPA or the U.S. Government. The U.S. Government is authorized to reproduce and distribute reprints for Governmental purposes notwithstanding any copyright annotation thereon. The work of H.G.K. was performed before joining Amazon Web Services.
Funding AgencyGrant Number
Office of the Director of National IntelligenceUNSPECIFIED
Air Force Office of Scientific Research (AFOSR)FA8721-05-C-0002
Intelligence Advanced Research Projects Activity (IARPA)IAA 1198
Intelligence Advanced Research Projects Activity (IARPA)IAA 8839, Annex 125
Issue or Number:3
Record Number:CaltechAUTHORS:20210927-213255608
Persistent URL:
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:111056
Deposited By: George Porter
Deposited On:27 Sep 2021 23:13
Last Modified:27 Sep 2021 23:13

Repository Staff Only: item control page