Abstract
Genetic Algorithm (GA) is a widely used metaheuristic for addressing challenging optimization problems. Selecting suitable settings for GA parameters can lead to a considerable improvement in its performance, but this is often difficult due to the larger number of alternatives available and variable performance depending on the problem solved. Furthermore, consideration of multiple performance criteria in parameter selection adds an additional layer of complexity. Therefore, practitioners often rely on their previous experiences or perform trial-and-error experiments when determining suitable parameter settings. In this study, we define the problem of finding suitable settings for GAs as a bi-objective optimization problem with an aim to find efficient settings that will maximize the approximation quality and minimize the run time of the GA. We conduct an empirical study on a GA that solves Traveling Salesperson Problem (TSP). Two evolutionary algorithms are utilized in a nested manner to address the bi-objective parameter setting problem: a GA to solve the TSP instances and a multi-objective evolutionary algorithm to find the bi-objective efficient settings of the GA. We find efficient settings for each instance through an empirical study with 31 TSP instances and identify settings that perform well regardless of the instance solved. The results provide valuable insights into the behavior of efficient settings, demonstrating that some operators and parameters are robust to changes in problem size, while others require adjustment.
Similar content being viewed by others
Data availability
The data that support the findings of this study are available at https://github.com/edasdemirlab/ai-ga-setting-code.git.
References
Coy SP, Golden BL, Runger GC, Wasil EA (2001) Using experimental design to find effective parameter settings for heuristics. J Heuristics 7(1):77–97
De Jong K (2007) Parameter setting in EAs: a 30 year perspective. In: Lobo FG, Lima CF, Michalewicz Z (eds) Parameter Setting in Evolutionary Algorithms. Springer, Berlin, Heidelberg, p 1–18
Huang CW, Li YX, Yao X (2020) A survey of automatic parameter tuning methods for metaheuristics. IEEE Trans Evol Comput 24(2):201–216
Ha QM, Deville Y, Pham QD, Ha MH (2020) A hybrid genetic algorithm for the traveling salesman problem with drone. J Heuristics 26(2):219–247
Zhang PL, Wang JQ, Tian ZW, Sun SZ, Li JT, Yang JN (2022) A genetic algorithm with jumping gene and heuristic operators for traveling salesman problem. Appl Soft Comput 127:109339
Panwar K, Deep K (2022) Discrete salp swarm algorithm for Euclidean travelling salesman problem. Appl Intell 53(10):11420–11438
Zheng RZ, Zhang Y, Yang K (2022) A transfer learning-based particle swarm optimization algorithm for travelling salesman problem. J Comput Des Eng 9(3):933–948
Yang K, You XM, Liu S, Pan H (2020) A novel ant colony optimization based on game for traveling salesman problem. Appl Intell 50(12):4529–4542
Mosayebi M, Sodhi M (2020) Tuning genetic algorithm parameters using design of experiments. In: Proceedings of the 2020 Genetic and Evolutionary Computation Conference Companion. Association for Computing Machinery, p 1937–1944
Ramos ICO, Goldbarg MC, Goldbarg EG, Neto ADD (2005) Logistic regression for parameter tuning on an evolutionary algorithm. 2005 IEEE Congress on Evolutionary Computation vols 1–3, p 1061–1068
Katoch S, Chauhan SS, Kumar V (2021) A review on genetic algorithm: past, present, and future. Multimedia Tools Appl 80(5):8091–8126
Smit SK, Eiben AE (2009) Comparing parameter tuning methods for evolutionary algorithms. 2009 IEEE Congress on Evolutionary Computation, vols 1–5, p 399–406
Eiben AE, Smit SK (2011) Parameter tuning for configuring and analyzing evolutionary algorithms. Swarm Evol Comput 1(1):19–31
de Lacerda MGP, Pessoa LFD, Neto FBD, Ludermir TB, Kuchen H (2021) A systematic literature review on general parameter control for evolutionary and swarm-based algorithms. Swarm Evol Comput 60:100777
Nannen V, Eiben AE (2006) A method for parameter calibration and relevance estimation in evolutionary-algorithms. Gecco 2006: Genetic and Evolutionary Computation Conference vols 1 and 2, 183
Bartz-Beielstein T, Lasarczyk CWG, Preuss M (2005) Sequential parameter optimization. In: 2005 IEEE Congress on Evolutionary Computation
Czarn A, MacNish C, Vijayan K, Turlach B, Gupta R (2004) Statistical exploratory analysis of genetic algorithms. IEEE Trans Evol Comput 8(4):405–421
Vasko FJ, Bobeck JD, Governale MA, Rieksts DJ, Keffer JD (2011) A statistical analysis of parameter values for the rank-based ant colony optimization algorithm for the traveling salesperson problem. J Oper Res Soc 62(6):1169–1176
Adenso-Diaz B, Laguna M (2006) Fine-tuning of algorithms using fractional experimental designs and local search. Oper Res 54(1):99–114
Mejia-de-Dios JA, Mezura-Montes E, Quiroz-Castellanos M (2021) Automated parameter tuning as a bilevel optimization problem solved by a surrogate-assisted population-based approach. Appl Intell 51(8):5978–6000
Grefenstette JJ (1986) Optimization of control parameters for genetic algorithms. IEEE Trans Syst Man Cybernetics 16(1):122–128
Dréo J (2009) Using performance fronts for parameter setting of stochastic metaheuristics. In: Proceedings of the 11th Annual Conference Companion on Genetic and Evolutionary Computation Conference: Late Breaking Papers. Association for Computing Machinery: Montreal, Québec, Canada, p 2197–2200
Ugolotti R, Cagnoni S (2014) Analysis of evolutionary algorithms using multi-objective parameter tuning. GECCO’14: Proceedings of the 2014 Genetic and Evolutionary Computation Conference, p 1343–1350
Ugolotti R, Sani L, Cagnoni S (2019) What can we learn from multi-objective meta-optimization of evolutionary algorithms in continuous domains? Mathematics 7(3):232
Zhang TT, Georgiopoulos M, Anagnostopoulos GC (2015) SPRINT Multi-Objective Model Racing. Gecco’15: Proceedings of the 2015 Genetic and Evolutionary Computation Conference, p 1383–1390
Zhang TT, Georgiopoulos M, Anagnostopoulos GC (2016) Multi-objective model selection via racing. IEEE Trans Cybernetics 46(8):1863–1876
Dasdemir E, Köksalan M, Tezcaner Öztürk D (2020) A flexible reference point-based multi-objective evolutionary algorithm: An application to the UAV route planning problem. Comput Oper Res 114:104811
Zhou A, Qu B-Y, Li H, Zhao S-Z, Suganthan PN, Zhang Q (2011) Multiobjective evolutionary algorithms: a survey of the state of the art. Swarm Evol Comput 1(1):32–49
Deb K (2015) Multi-objective evolutionary algorithms. In: Kacprzyk J, Pedrycz W (eds) Springer Handbook of Computational Intelligence. Springer, Berlin Heidelberg, pp 995–1015
GUROBI (2023) Documentation: MIPGap; Available from: https://www.gurobi.com/documentation/9.0/refman/mipgap.html. Accessed 1 May 2023
Deb K, Pratap A, Agarwal S, Meyarivan T (2002) A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Trans Evol Comput 6(2):182–197
Scrucca L (2013) GA: a package for genetic algorithms in R. J Stat Softw 53:1–37
R Core Team (2023) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available from: https://www.R-project.org/. Accessed 1 May 2023
Paul PV, Ganeshkumar C, Dhavachelvan P, Baskaran R (2020) A novel ODV crossover operator-based genetic algorithms for traveling salesman problem. Soft Comput 24(17):12855–12885
CCR (2023) The Center for computational research. University at Buffalo. http://hdl.handle.net/10477/79221. Accessed 1 May 2023
Tsou CS (2013) nsga2R: elitist non-dominated sorting genetic algorithm. R Package version, 1.1. Available from: https://CRAN.R-project.org/package=nsga2R. Accessed 1 May 2023
Mersmann O, Trautmann H, Steuer D, Bischl B, Deb K (2014) mco: multiple criteria optimization algorithms and related functions. R package version, 1.15.6. Available from: https://CRAN.R-project.org/package=mco. Accessed 1 May 2023
Nagata Y, Kobayashi S (2013) A powerful genetic algorithm using edge assembly crossover for the traveling salesman problem. Informs J Comput 25(2):346–363
Acknowledgements
This paper is based on the MSc thesis of the first author, co-supervised by the remaining authors. We are grateful to the University at Buffalo for allowing us to use their Center for Computational Research for high-performance computing resources.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest/ Competing interests
None.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix
Supplementary File
(DOCX 45 KB)
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Akduran, Y., Dasdemir, E. & Testik, M.C. Bi-objective parameter setting problem of a genetic algorithm: an empirical study on traveling salesperson problem. Appl Intell 53, 27148–27162 (2023). https://doi.org/10.1007/s10489-023-04972-z
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10489-023-04972-z