ABSTRACT
The design of effective features enabling the development of automated landscape-aware techniques requires to address a number of inter-dependent issues. In this paper, we are interested in contrasting the amount of budget devoted to the computation of features with respect to: (i) the effectiveness of the features in grasping the characteristics of the landscape, and (ii) the gain in accuracy when solving an unknown problem instance by means of a feature-informed automated algorithm selection approach. We consider multi-objective combinatorial landscapes where, to the best of our knowledge, no in depth investigations have been conducted so far. We study simple cost-adjustable sampling strategies for extracting different state-of-the-art features. Based on extensive experiments, we report a comprehensive analysis on the impact of sampling on landscape feature values, and the subsequent automated algorithm selection task. In particular, we identify different global trends of feature values leading to non-trivial cost-vs-accuracy trade-off(s). Besides, we provide evidence that the sampling strategy can improve the prediction accuracy of automated algorithm selection. Interestingly, this holds independently of whether the sampling cost is taken into account or not in the overall solving budget.
- Hernan Aguirre and Kiyoshi Tanaka. 2007. Working principles, behavior, and performance of MOEAs on MNK-landscapes. European Journal of Operational Research 181 (09 2007), 1670--1690. Google ScholarCross Ref
- Leo Breiman. 2001. Random Forests. Machine Learning 45, 1 (2001), 5--32.Google ScholarDigital Library
- Cosson, Derbel, Liefooghe, Aguirre, Tanaka, and Zhang. 2021. Decomposition-Based Multi-objective Landscape Features and Automated Algorithm Selection. In Evolutionary Computation in Combinatorial Optimization, Christine Zarges and Sébastien Verel (Eds.). Springer International Publishing, 34--50.Google Scholar
- K. Deb. 2001. Multi-Objective Optimization using Evolutionary Algorithms. Cambridge University Press. 409--410 pages.Google ScholarDigital Library
- Matthias Ehrgott. 2005. Multicriteria Optimization (2. ed.). Springer. Google ScholarCross Ref
- Deon Garrett and Dipankar Dasgupta. 2007. Multiobjective Landscape Analysis and the Generalized Assignment Problem. 110--124. Google ScholarDigital Library
- S. A. Kauffman. 1993. The Origins of Order. Oxford University Press.Google Scholar
- Pascal Kerschke, Holger H. Hoos, Frank Neumann, and Heike Trautmann. 2019. Automated Algorithm Selection: Survey and Perspectives. Evol. Comput. 27, 1 (2019), 3--45.Google ScholarDigital Library
- Pascal Kerschke, Mike Preuss, Simon Wessing, and Heike Trautmann. 2016. Low-Budget Exploratory Landscape Analysis on Multiple Peaks Models. In The on Genetic and Evolutionary Computation Conference GECCO. ACM, 229--236. Google ScholarDigital Library
- Pascal Kerschke and Heike Trautmann. 2019. Automated Algorithm Selection on Continuous Black-Box Problems by Combining Exploratory Landscape Analysis and Machine Learning. Evol. Comput. 27, 1 (2019), 99--127. Google ScholarDigital Library
- Joshua Knowles and David Corne. 2003. Instance Generators and Test Suites for the Multiobjective Quadratic Assignment Problem. In Evolutionary Multi-Criterion Optimization, Carlos M. Fonseca, Peter J. Fleming, Eckart Zitzler, Lothar Thiele, and Kalyanmoy Deb (Eds.). Springer Berlin Heidelberg, Berlin, Heidelberg, 295--310.Google Scholar
- Arnaud Liefooghe, Fabio Daolio, Sébastien Verel, Bilel Derbel, Hernan Aguirre, and Kiyoshi Tanaka. 2020. Landscape-aware performance prediction for evolutionary multi-objective optimization. IEEE Transactions on Evolutionary Computation (2020).Google Scholar
- Arnaud Liefooghe, Sébastien Vérel, Bilel Derbel, Hernán E. Aguirre, and Kiyoshi Tanaka. 2020. Dominance, Indicator and Decomposition Based Search for Multi-objective QAP: Landscape Analysis and Automated Algorithm Selection. In Parallel Problem Solving from Nature - PPSN XVI - 16th International Conference, PPSN 2020, Leiden, The Netherlands, September 5-9, 2020, Proceedings, Part I (Lecture Notes in Computer Science, Vol. 12269), Thomas Bäck, Mike Preuss, André H. Deutz, Hao Wang, Carola Doerr, Michael T. M. Emmerich, and Heike Trautmann (Eds.). Springer, 33--47. Google ScholarDigital Library
- Arnaud Liefooghe, Sébastien Vérel, Benjamin Lacroix, Alexandru-Ciprian Zavoianu, and John A. W. McCall. 2021. Landscape features and automated algorithm selection for multi-objective interpolated continuous optimisation problems. In Genetic and Evolutionary Computation (GECCO 2021). ACM, Lille, France, 421--429. Google ScholarDigital Library
- Marius Lindauer, Jan N. van Rijn, and Lars Kotthoff. 2018. The Algorithm Selection Competition Series 2015-17. CoRR abs/1805.01214 (2018). arXiv:1805.01214Google Scholar
- Katherine Mary Malan. 2021. A Survey of Advances in Landscape Analysis for Optimisation. Algorithms 14, 2 (2021), 40. Google ScholarCross Ref
- Gauvain Marquet, Bilel Derbel, Arnaud Liefooghe, and El-Ghazali Talbi. 2014. Shake Them All! - Rethinking Selection and Replacement in MOEA/D. In PPSN XIII. 641--651.Google Scholar
- Olaf Mersmann, Bernd Bischl, Heike Trautmann, Mike Preuss, Claus Weihs, and Günter Rudolph. 2011. Exploratory Landscape Analysis. In GECCO. 829--836.Google Scholar
- Rachael Morgan and Marcus Gallagher. 2014. Sampling Techniques and Distance Metrics in High Dimensional Continuous Landscape Analysis: Limitations and Improvements. IEEE Transactions on Evolutionary Computation 18, 3 (2014), 456--461. Google ScholarCross Ref
- Luis Paquete, Tommaso Schiavinotto, and Thomas Stützle. 2007. On local optima in multiobjective combinatorial optimization problems. Annals of Operations Research 156 (09 2007), 83--97. Google ScholarCross Ref
- Fabian Pedregosa, Gaël Varoquaux, Alexandre Gramfort, Vincent Michel, Bertrand Thirion, Olivier Grisel, Mathieu Blondel, Peter Prettenhofer, Ron Weiss, Vincent Dubourg, et al. 2011. Scikit-learn: Machine learning in Python. Journal of machine learning research 12, Oct (2011), 2825--2830.Google ScholarDigital Library
- Quentin Renau, Carola Doerr, Johann Dréo, and Benjamin Doerr. 2020. Exploratory Landscape Analysis is Strongly Sensitive to the Sampling Strategy. In Parallel Problem Solving from Nature - PPSN XVI - 16th International Conference, PPSN (Lecture Notes in Computer Science, Vol. 12270). Springer, 139--153. Google ScholarDigital Library
- Quentin Renau, Johann Dréo, Carola Doerr, and Benjamin Doerr. 2019. Expressiveness and robustness of landscape features. In The Genetic and Evolutionary Computation Conference Companion, GECCO. ACM, 2048--2051. Google ScholarDigital Library
- John R. Rice. 1976. The Algorithm Selection Problem. Adv Comput 15 (1976), 65--118.Google ScholarCross Ref
- Hendrik Richter and Andries Engelbrecht (Eds.). 2014. Recent Advances in the Theory and Application of Fitness Landscapes. Springer.Google Scholar
- Sobia Saleem, Marcus Gallagher, and Ian Wood. 2019. Direct Feature Evaluation in Black-Box Optimization Using Problem Transformations. Evolutionary Computation 27, 1 (03 2019), 75--98. arXiv:https://direct.mit.edu/evco/article-pdf/27/1/75/1553218/evco_a_00247.pdf Google ScholarDigital Library
- Sarah L. Thomson, Gabriela Ochoa, and Sébastien Vérel. 2019. Clarifying the Difference in Local Optima Network Sampling Algorithms. In Evolutionary Computation in Combinatorial Optimization - 19th European Conference, EvoCOP (Lecture Notes in Computer Science, Vol. 11452). Springer, 163--178. Google ScholarDigital Library
- Sarah L. Thomson, Gabriela Ochoa, Sébastien Vérel, and Nadarajen Veerapen. 2020. Inferring Future Landscapes: Sampling the Local Optima Level. Evol. Comput. 28, 4 (2020), 621--641. Google ScholarDigital Library
- Leonardo Vanneschi, Sébastien Vérel, Marco Tomassini, and Philippe Collard. 2009. NK Landscapes Difficulty and Negative Slope Coefficient: How Sampling Influences the Results. In Applications of Evolutionary Computing, EvoWorkshops (Lecture Notes in Computer Science, Vol. 5484). Springer, 645--654. Google ScholarCross Ref
- Sébastien Verel, Arnaud Liefooghe, Laetitia Jourdan, and Clarisse Dhaenens. 2013. On the structure of multiobjective combinatorial search space: MNK-landscapes with correlated objectives. European Journal of Operational Research 227, 2 (2013), 331--342.Google ScholarCross Ref
- E. D. Weinberger. 1990. Correlated and uncorrelatated fitness landscapes and how to tell the difference. Biol Cybern 63, 5 (1990), 325--336.Google ScholarDigital Library
- Qingfu Zhang and Hui Li. 2008. MOEA/D: A Multiobjective Evolutionary Algorithm Based on Decomposition. IEEE TEVC 11 (01 2008), 712 -- 731.Google Scholar
- E. Zitzler, L. Thiele, M. Laumanns, C. M. Fonseca, and V. Grunert da Fonseca. 2003. Performance Assessment of Multiobjective Optimizers: An Analysis and Review. IEEE TEVC 7, 2 (2003), 117--132.Google Scholar
Index Terms
- Cost-vs-accuracy of sampling in multi-objective combinatorial exploratory landscape analysis
Recommendations
Multi-objective NK landscapes with heterogeneous objectives
GECCO '22: Proceedings of the Genetic and Evolutionary Computation ConferenceSo far, multi-objective NK landscapes have been investigated under the assumption of a homogeneous nature of the involved objectives in terms of difficulty. However, we argue that problems with heterogeneous objectives, e.g., in terms of multi-modality, ...
Covariance Matrix Adaptation for Multi-objective Optimization
The covariancematrix adaptation evolution strategy (CMA-ES) is one of themost powerful evolutionary algorithms for real-valued single-objective optimization. In this paper, we develop a variant of the CMA-ES for multi-objective optimization (MOO). We ...
An analysis on recombination in multi-objective evolutionary optimization
Evolutionary algorithms (EAs) are increasingly popular approaches to multi-objective optimization. One of their significant advantages is that they can directly optimize the Pareto front by evolving a population of solutions, where the recombination (...
Comments