ABSTRACT
Specialized graybox local search and crossover have been successfully combined within the framework of the so-called Drils (Deterministic recombination and iterated local search) algorithm. As for any evolutionary algorithm, the initial design framework, and the underlying high-level choices and parameters, are crucially important. The Drils algorithm is no exception, and recent enhanced variants exist in the literature. In this paper, we aim at: (i) improving the performance of the latest variants of Drils, and (ii) providing a better principled understanding of graybox search behavior and dynamics. On the basis of a preliminary analysis using Local Optima Networks of small-size NKQ-landscapes, we first highlight the difference of using local search with and without crossover. We then propose to pipeline these two techniques in a simple two-phase like iterated local search scheme which is shown to provide substantial improvements over the latest Drils+ variant for large-size NKQ-landscapes. We further report a dedicated analysis in an attempt to provide new insights into the impact of local search and crossover on the phenotype and the genotype of the local optima encountered in the search trajectory.
- Lorenzo Canonne and Bilel Derbel. 2022. Drils revisited: on the combination of perturbation with graybox optimization techniques. In GECCO '22: Genetic and Evolutionary Computation Conference, Boston, Massachusetts, USA, July 9 -- 13, 2022, Jonathan E. Fieldsend and Markus Wagner (Eds.). ACM, 204--212. Google ScholarDigital Library
- Wenxiang Chen, L. Darrell Whitley, Doug Hains, and Adele E. Howe. 2013. Second order partial derivatives for NK-landscapes. In Genetic and Evolutionary Computation Conference, GECCO '13, Amsterdam, The Netherlands, July 6--10, 2013, Christian Blum and Enrique Alba (Eds.). ACM, 503--510. Google ScholarDigital Library
- Francisco Chicano, Gabriela Ochoa, L. Darrell Whitley, and Renato Tinós. 2018. Enhancing partition crossover with articulation points analysis. In Proceedings of the Genetic and Evolutionary Computation Conference, GECCO 2018, Kyoto, Japan, July 15--19, 2018, Hernán E. Aguirre and Keiki Takadama (Eds.). ACM, 269--276. Google ScholarDigital Library
- Francisco Chicano, Gabriela Ochoa, L. Darrell Whitley, and Renato Tinós. 2019. Quasi-Optimal Recombination Operator. In Evolutionary Computation in Combinatorial Optimization - 19th European Conference, EvoCOP 2019, Held as Part of EvoStar 2019, Leipzig, Germany, April 24--26, 2019, Proceedings (Lecture Notes in Computer Science, Vol. 11452), Arnaud Liefooghe and Luís Paquete (Eds.). Springer, 131--146. Google ScholarDigital Library
- Francisco Chicano, Darrell Whitley, and Andrew M. Sutton. 2014. Efficient Identification of Improving Moves in a Ball for Pseudo-Boolean Problems. In Proceedings of the 2014 Annual Conference on Genetic and Evolutionary Computation (Vancouver, BC, Canada) (GECCO '14). Association for Computing Machinery, New York, NY, USA, 437--444. Google ScholarDigital Library
- Francisco Chicano, L. Darrell Whitley, Gabriela Ochoa, and Renato Tinós. 2017. Optimizing one million variable NK landscapes by hybridizing deterministic recombination and local search. In Proceedings of the Genetic and Evolutionary Computation Conference, GECCO 2017, Berlin, Germany, July 15--19, 2017, Peter A. N. Bosman (Ed.). ACM, 753--760. Google ScholarDigital Library
- Brian W. Goldman and William F. Punch. 2015. Gray-Box Optimization using the Parameter-less Population Pyramid. In Proceedings of the Genetic and Evolutionary Computation Conference, GECCO 2015, Madrid, Spain, July 11--15, 2015, Juan Luis Jiménez Laredo, Sara Silva, and Anna Isabel Esparcia-Alcázar (Eds.). ACM, 855--862. Google ScholarDigital Library
- Holger H. Hoos and Thomas Stützle. 2004. Stochastic Local Search: Foundations & Applications. Elsevier / Morgan Kaufmann.Google Scholar
- Sungmin Hwang, Benjamin Schmiegelt, Luca Ferretti, and Joachim Krug. 2018. Universality Classes of Interaction Structures for NK Fitness Landscapes. Journal of Statistical Physics 172, 1 (2018), 226--278. Google ScholarCross Ref
- Stuart Kauffman and Simon Levin. 1987. Towards a general theory of adaptive walks on rugged landscapes. Journal of Theoretical Biology 128, 1 (1987), 11--45. Google ScholarCross Ref
- S. A. Kauffman. 1993. The Origins of Order. Oxford University Press.Google Scholar
- Florian Leprêtre, Sébastien Verel, Cyril Fonlupt, and Virginie Marion. 2019. Walsh Functions as Surrogate Model for Pseudo-Boolean Optimization Problems. In Proceedings of the Genetic and Evolutionary Computation Conference on - GECCO '19. ACM Press, Prague, Czech Republic, 303--311.Google ScholarDigital Library
- Helena Ramalhinho Lourenço, Olivier C. Martin, and Thomas Stützle. 2019. Iterated Local Search: Framework and Applications. Springer International Publishing, Cham, 129--168. Google ScholarCross Ref
- Katherine M. Malan. 2021. A Survey of Advances in Landscape Analysis for Optimisation. Algorithms 14, 2 (2021), 40. Google ScholarCross Ref
- Stefan Nowak and Joachim Krug. 2015. Analysis of adaptive walks on NK fitness landscapes with different interaction schemes. Journal of Statistical Mechanics: Theory and Experiment 2015, 6 (jun 2015), P06014. Google ScholarCross Ref
- G. Ochoa, M. Tomassini, S. Verel, and C. Darabos. 2008. A Study of NK Landscapes' Basins and Local Optima Networks. In Proceedings of the 10th Annual Conference on Genetic and Evolutionary Computation (Atlanta, GA, USA) (GECCO '08). ACM, New York, NY, USA, 555--562. Google ScholarDigital Library
- Gabriela Ochoa, Sébastien Verel, Fabio Daolio, and Marco Tomassini. 2014. Local Optima Networks: A New Model of Combinatorial Fitness Landscapes. Springer Berlin Heidelberg, Berlin, Heidelberg, 233--262. Google ScholarCross Ref
- Hendrik Richter and Andries Engelbrecht (Eds.). 2014. Recent Advances in the Theory and Application of Fitness Landscapes. Springer.Google Scholar
- Thomas Stützle and Rubén Ruiz. 2018. Iterated Local Search. In Handbook of Heuristics, Rafael Martí, Panos M. Pardalos, and Mauricio G. C. Resende (Eds.). Springer, 579--605.Google Scholar
- Audrey Terras. 1999. Fourier Analysis on Finite Groups and Applications. Cambridge University Press. Google ScholarCross Ref
- Renato Tinós, Michal W. Przewozniczek, and Darrell Whitley. 2022. Iterated Local Search with Perturbation Based on Variables Interaction for Pseudo-Boolean Optimization. In Proceedings of the Genetic and Evolutionary Computation Conference (Boston, Massachusetts) (GECCO '22). Association for Computing Machinery, New York, NY, USA, 296--304. Google ScholarDigital Library
- Renato Tinós, L. Darrell Whitley, and Francisco Chicano. 2015. Partition Crossover for Pseudo-Boolean Optimization. In Proceedings of the 2015 ACM Conference on Foundations of Genetic Algorithms XIII, Aberystwyth, United Kingdom, January 17 -- 20, 2015, Jun He, Thomas Jansen, Gabriela Ochoa, and Christine Zarges (Eds.). ACM, 137--149. Google ScholarDigital Library
- Darrell Whitley, Doug Hains, and Adele Howe. 2009. Tunneling between Optima: Partition Crossover for the Traveling Salesman Problem. In Proceedings of the 11th Annual Conference on Genetic and Evolutionary Computation (Montreal, Québec, Canada) (GECCO '09). Association for Computing Machinery, New York, NY, USA, 915--922. Google ScholarDigital Library
- L. Darrell Whitley and Wenxiang Chen. 2012. Constant time steepest descent local search with lookahead for NK-landscapes and MAX-kSAT. In Genetic and Evolutionary Computation Conference, GECCO '12, Philadelphia, PA, USA, July 7--11, 2012, Terence Soule and Jason H. Moore (Eds.). ACM, 1357--1364. Google ScholarDigital Library
- L. Darrell Whitley, Francisco Chicano, and Brian W. Goldman. 2016. Gray Box Optimization for Mk Landscapes Nk Landscapes and Max-Ksat. Evol. Comput. 24, 3 (Sept. 2016), 491--519. Google ScholarDigital Library
- Feng Zou, Debao Chen, Hui Liu, Siyu Cao, Xuying Ji, and Yan Zhang. 2022. A survey of fitness landscape analysis for optimization. Neurocomputing 503 (2022), 129--139. Google ScholarDigital Library
Index Terms
- To Combine or not to Combine Graybox Crossover and Local Search?
Recommendations
Multi-layer local optima networks for the analysis of advanced local search-based algorithms
GECCO '20: Proceedings of the 2020 Genetic and Evolutionary Computation ConferenceA Local Optima Network (LON) is a graph model that compresses the fitness landscape of a particular combinatorial optimization problem based on a specific neighborhood operator and a local search algorithm. Determining which and how landscape features ...
Bag of local landscape features for fitness landscape analysis
Evolutionary computation is a research field dealing with black-box and complex optimization problems whose fitness landscapes are usually unknown in advance. It is difficult to select an appropriate evolutionary algorithm and parameters for a given ...
Fitness Landscape Analysis of Genetic Programming Search Spaces with Local Optima Networks
GECCO '23 Companion: Proceedings of the Companion Conference on Genetic and Evolutionary ComputationFitness landscape analysis (FLA) refers to a set of techniques to characterise optimisation problems. This paper presents an FLA of three types of genetic programming (GP) benchmarks: parity, symbolic regression, and artificial ant. We applied a ...
Comments