ABSTRACT
The exploration vs exploitation dilemma is to balance exploring new but potentially less fit regions of the fitness landscape while also focusing on regions near the fittest individuals. For the tunable problem class SparseLocalOpt, a non-elitist EA with tournament selection can limit the percentage of "sparse" local optimal individuals in the population using a sufficiently high mutation rate (Dang et al., 2021). However, the performance of the EA depends critically on choosing the "right" mutation rate, which is problem instance-specific. A promising approach is self-adaptation, where parameter settings are encoded in chromosomes and evolved.
We propose a new self-adaptive EA for single-objective optimisation, which treats parameter control from the perspective of multiobjective optimisation: The algorithm simultaneously maximises the fitness and the mutation rates. Since individuals in "dense" fitness valleys survive high mutation rates, and individuals on "sparse" local optima only survive with lower mutation rates, they can coexist on a non-dominated Pareto front.
Runtime analyses show that this new algorithm (MOSA-EA) can efficiently escape a local optimum with unknown sparsity, where some fixed mutation rate EAs become trapped. Complementary experimental results show that the MOSA-EA outperforms a range of EAs on random NK-Landscape and k-Sat instances.
Supplemental Material
Available for Download
Supplemental material.
- Dimitris Achlioptas and Cristopher Moore. 2006. Random k-SAT: Two Moments Sufice to Cross a Sharp Threshold. SIAM J. Comput. 36, 3 (Jan. 2006), 740--762.Google ScholarDigital Library
- Denis Antipov, Benjamin Doerr, and Vitalii Karavaev. 2022. A Rigorous Runtime Analysis of the (1 + (λ, λ)) GA on Jump Functions. Algorithmica (Jan. 2022).Google Scholar
- Henry Bambury, Antoine Bultel, and Benjamin Doerr. 2021. Generalized jump functions. In Proceedings of the Genetic and Evolutionary Computation Conference. ACM, Lille France, 1124--1132.Google ScholarDigital Library
- Guy Bresler and Brice Huang. 2021. The Algorithmic Phase Transition of Random k-SAT for Low Degree Polynomials. arXiv:2106.02129 [math-ph, stat] (Oct. 2021). arXiv: 2106.02129.Google Scholar
- Brendan Case and Per Kristian Lehre. 2020. Self-adaptation in non-Elitist Evolutionary Algorithms on Discrete Problems with Unknown Structure. IEEE Transactions on Evolutionary Computation (2020), 1--1.Google Scholar
- Amin Coja-Oghlan. 2014. The asymptotic k-SAT threshold. In Proceedings of the forty-sixth annual ACM symposium on Theory of computing. ACM, New York New York, 804--813.Google ScholarDigital Library
- Dogan Corus, Duc-Cuong Dang, Anton V. Eremeev, and Per Kristian Lehre. 2018. Level-Based Analysis of Genetic Algorithms and Other Search Processes. IEEE Transactions on Evolutionary Computation 22, 5 (Oct. 2018), 707--719.Google ScholarDigital Library
- Duc-Cuong Dang, Anton Eremeev, and Per Kristian Lehre. 2020. Escaping Local Optima with Non-Elitist Evolutionary Algorithms. In Proceedings of AAAI 2021. AAAI Press.Google Scholar
- Duc-Cuong Dang, Anton Eremeev, and Per Kristian Lehre. 2021. Non-elitist Evolutionary Algorithms Excel in Fitness Landscapes with Sparse Deceptive Regions and Dense Valleys. In Proceedings of the Genetic and Evolutionary Computation Conference. ACM, Lille, France.Google ScholarDigital Library
- Duc-Cuong Dang, Tobias Friedrich, Timo Kotzing, Martin S. Krejca, Per Kristian Lehre, Pietro S. Oliveto, Dirk Sudholt, and Andrew M. Sutton. 2018. Escaping Local Optima Using Crossover With Emergent Diversity. IEEE Transactions on Evolutionary Computation 22, 3 (June 2018), 484--497.Google ScholarCross Ref
- Duc-Cuong Dang and Per Kristian Lehre. 2016. Runtime Analysis of Non-elitist Populations: From Classical Optimisation to Partial Information. Algorithmica 75, 3 (July 2016), 428--461.Google ScholarDigital Library
- Duc-Cuong Dang and Per Kristian Lehre. 2016. Self-adaptation of Mutation Rates in Non-elitist Populations. In Parallel Problem Solving from Nature - PPSN XIV. Vol. 9921. Springer International Publishing, Cham, 803--813.Google Scholar
- Duc-Cuong Dang, Per Kristian Lehre, and Phan Trung Hai Nguyen. 2019. Level-Based Analysis of the Univariate Marginal Distribution Algorithm. Algorithmica 81, 2 (Feb. 2019), 668--702.Google ScholarDigital Library
- K. Deb, A. Pratap, S. Agarwal, and T. Meyarivan. 2002. A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Transactions on Evolutionary Computation 6, 2 (April 2002), 182--197.Google ScholarDigital Library
- Benjamin Doerr. 2022. Does Comma Selection Help to Cope with Local Optima? Algorithmica (Jan. 2022).Google Scholar
- Benjamin Doerr and Timo Kötzing. 2020. Multiplicative Up-Drift. Algorithmica (Oct. 2020).Google Scholar
- Benjamin Doerr, Carsten Witt, and Jing Yang. 2021. Runtime Analysis for Selfadaptive Mutation Rates. Algorithmica 83, 4 (April 2021), 1012--1053.Google ScholarDigital Library
- Jens Gottlieb, Elena Marchiori, and Claudio Rossi. 2002. Evolutionary Algorithms for the Satisfiability Problem. Evolutionary Computation 10, 1 (March 2002), 35--50.Google ScholarDigital Library
- Mario Alejandro Hevia Fajardo and Dirk Sudholt. 2021. Self-adjusting offspring population sizes outperform fixed parameters on the cliff function. In Proceedings of the 16th ACM/SIGEVO Conference on Foundations of Genetic Algorithms. ACM, Virtual Event Austria, 1--15.Google ScholarDigital Library
- Stuart A Kauffman and Edward D Weinberger. 1989. The NK model of rugged fitness landscapes and its application to maturation of the immune response. Journal of theoretical biology 141, 2 (1989), 211--245. Publisher: Elsevier.Google ScholarCross Ref
- Per Kristian Lehre. 2010. Negative Drift in Populations. In Parallel Problem Solving from Nature, PPSN XI. Springer Berlin Heidelberg, Berlin, Heidelberg, 244--253.Google Scholar
- Per Kristian Lehre and Xiaoyu Qin. 2021. More Precise Runtime Analyses of Non-elitist EAs in Uncertain Environments. In Proceedings of the Genetic and Evolutionary Computation Conference. ACM, Lille, France, 9.Google ScholarDigital Library
- Per Kristian Lehre and Xin Yao. 2012. On the Impact of Mutation-Selection Balance on the Runtime of Evolutionary Algorithms. IEEE Transactions on Evolutionary Computation 16, 2 (April 2012), 225--241.Google ScholarDigital Library
- Heinz Mühlenbein and Gerhard Paaß. 1996. From recombination of genes to the estimation of distributions I. Binary parameters. In International conference on parallel problem solving from nature. Springer, 178--187.Google ScholarDigital Library
- Constantin P. Niculescu and Andrei Vernescu. 2004. A two sided estimate of ex - (1 + x/n)n. Journal of Inequalities in Pure and Applied Mathematics 5, 3 (2004).Google Scholar
- Amirhossein Rajabi and Carsten Witt. 2021. Stagnation detection in highly multimodal fitness landscapes. In Proceedings of the Genetic and Evolutionary Computation Conference. ACM, Lille France, 1178--1186.Google ScholarDigital Library
- N. Srinivas and Kalyanmoy Deb. 1994. Muiltiobjective Optimization Using Nondominated Sorting in Genetic Algorithms. Evolutionary Computation 2, 3 (Sept. 1994), 221--248.Google ScholarDigital Library
- Matej Črepinšek, Shih-Hsi Liu, and Marjan Mernik. 2013. Exploration and exploitation in evolutionary algorithms: A survey. Comput. Surveys 45, 3 (June 2013), 1--33.Google ScholarDigital Library
Index Terms
- Self-adaptation via multi-objectivisation: a theoretical study
Recommendations
Self-adaptation via Multi-objectivisation: An Empirical Study
Parallel Problem Solving from Nature – PPSN XVIIAbstractNon-elitist evolutionary algorithms (EAs) can be beneficial in optimisation of noisy and or rugged fitness landscapes. However, this benefit can only be realised if the parameters of the non-elitist EAs are carefully adjusted in accordance with ...
Use of the q-Gaussian mutation in evolutionary algorithms
Special issue on advances in computational intelligence and bioinformaticsThis paper proposes the use of the q-Gaussian mutation with self-adaptation of the shape of the mutation distribution in evolutionary algorithms. The shape of the q-Gaussian mutation distribution is controlled by a real parameter q. In the proposed ...
Self-adaptation Can Help Evolutionary Algorithms Track Dynamic Optima
GECCO '23: Proceedings of the Genetic and Evolutionary Computation ConferenceReal-world optimisation problems often involve dynamics, where objective functions may change over time. Previous studies have shown that evolutionary algorithms (EAs) can solve dynamic optimisation problems. Additionally, the use of diversity ...
Comments