Parameterless Gene-pool Optimal Mixing Evolutionary Algorithms

When it comes to solving optimization problems with evolutionary algorithms (EAs) in a reliable and scalable manner, detecting and exploiting linkage information, i.e., dependencies between variables, can be key. In this article, we present the latest version of, and propose substantial enhancements to, the Gene-pool Optimal Mixing Evoutionary Algorithm (GOMEA): an EA explicitly designed to estimate and exploit linkage information. We begin by performing a large-scale search over several GOMEA design choices, to understand what matters most and obtain a generally bestperforming version of the algorithm. Next, we introduce a novel version of GOMEA, called CGOMEA, where linkage-based variation is further improved by filtering solution mating based on conditional dependencies. We compare our latest version of GOMEA, the newly introduced CGOMEA, and another contending linkage-aware EA DSMGAII in an extensive experimental evaluation, involving a benchmark set of 9 black-box problems that can only be solved efficiently if their inherent dependency structure is unveiled and exploited. Finally, in an attempt to make EAs more usable and resilient to parameter choices, we investigate the performance of different automatic population management schemes for GOMEA and CGOMEA, de facto making the EAs parameterless. Our results show that GOMEA and CGOMEA significantly outperform the original GOMEA and DSMGA-II on most problems, setting a new state of the art for the field.

[1]  Kalyanmoy Deb,et al.  Analyzing Deception in Trap Functions , 1992, FOGA.

[2]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[3]  Kalyanmoy Deb,et al.  Multimodal Deceptive Functions , 1993, Complex Syst..

[4]  Rich Caruana,et al.  Removing the Genetics from the Standard Genetic Algorithm , 1995, ICML.

[5]  Paul A. Viola,et al.  MIMIC: Finding Optima by Estimating Probability Densities , 1996, NIPS.

[6]  Hillol Kargupta,et al.  The Gene Expression Messy Genetic Algorithm , 1996, Proceedings of IEEE International Conference on Evolutionary Computation.

[7]  David H. Wolpert,et al.  No free lunch theorems for optimization , 1997, IEEE Trans. Evol. Comput..

[8]  S. Baluja,et al.  Combining Multiple Optimization Runs with Optimal Dependency Trees , 1997 .

[9]  Heinz Mühlenbein,et al.  FDA -A Scalable Evolutionary Algorithm for the Optimization of Additively Decomposed Functions , 1999, Evolutionary Computation.

[10]  Fernando G. Lobo,et al.  A parameter-less genetic algorithm , 1999, GECCO.

[11]  David E. Goldberg,et al.  The compact genetic algorithm , 1999, IEEE Trans. Evol. Comput..

[12]  D. Goldberg,et al.  BOA: the Bayesian optimization algorithm , 1999 .

[13]  M. Pelikán,et al.  The Bivariate Marginal Distribution Algorithm , 1999 .

[14]  Dirk Thierens,et al.  Scalability Problems of Simple Genetic Algorithms , 1999, Evolutionary Computation.

[15]  J. A. Lozano,et al.  Estimation of Distribution Algorithms: A New Tool for Evolutionary Computation , 2001 .

[16]  Martin Pelikan,et al.  Bayesian Optimization Algorithm , 2005 .

[17]  William E. Hart,et al.  Memetic Evolutionary Algorithms , 2005 .

[18]  David E. Goldberg,et al.  Hierarchical Bayesian Optimization Algorithm , 2006, Scalable Optimization via Probabilistic Modeling.

[19]  Kumara Sastry,et al.  Linkage Learning via Probabilistic Modeling in the Extended Compact Genetic Algorithm (ECGA) , 2006, Scalable Optimization via Probabilistic Modeling.

[20]  Pedro Larrañaga,et al.  Towards a New Evolutionary Computation - Advances in the Estimation of Distribution Algorithms , 2006, Towards a New Evolutionary Computation.

[21]  Alexander Kraskov,et al.  MIC: Mutual Information Based Hierarchical Clustering , 2008, 0809.1605.

[22]  Martin V. Butz,et al.  Performance of evolutionary algorithms on NK landscapes with nearest neighbor interactions and tunable overlap , 2009, GECCO '09.

[23]  Dirk Thierens,et al.  The roles of local search, model building and optimal mixing in evolutionary algorithms from a bbo perspective , 2011, GECCO.

[24]  Dirk Thierens,et al.  Optimal mixing evolutionary algorithms , 2011, GECCO '11.

[25]  Dirk Thierens,et al.  Linkage neighbors, optimal mixing and forced improvements in genetic algorithms , 2012, GECCO '12.

[26]  Dirk Thierens,et al.  On Measures to Build Linkage Trees in LTGA , 2012, PPSN.

[27]  Dirk Thierens,et al.  More concise and robust linkage learning by filtering and combining linkage hierarchies , 2013, GECCO '13.

[28]  William F. Punch,et al.  Fast and Efficient Black Box Optimization Using the Parameter-less Population Pyramid , 2015, Evolutionary Computation.

[29]  Tian-Li Yu,et al.  Optimization by Pairwise Linkage Detection, Incremental Linkage Set, and Restricted / Back Mixing: DSMGA-II , 2015, GECCO.

[30]  Dirk Thierens,et al.  The Multiple Insertion Pyramid: A Fast Parameter-Less Population Scheme , 2016, PPSN.

[31]  Peter A. N. Bosman,et al.  Scalable genetic programming by gene-pool optimal mixing and input-space entropy-based building-block learning , 2017, GECCO.

[32]  Tian-Li Yu,et al.  Two-edge graphical linkage model for DSMGA-II , 2017, GECCO.

[33]  L. Darrell Whitley,et al.  Optimizing one million variable NK landscapes by hybridizing deterministic recombination and local search , 2017, GECCO.

[34]  Tian-Li Yu,et al.  Investigation of the exponential population scheme for genetic algorithms , 2018, GECCO.

[35]  Peter A. N. Bosman,et al.  Multi-objective Gene-pool Optimal Mixing Evolutionary Algorithm with the Interleaved Multi-start Scheme , 2018, Swarm Evol. Comput..

[36]  Peter A. N. Bosman,et al.  Symbolic regression and feature construction with GP-GOMEA applied to radiotherapy dose reconstruction of childhood cancer survivors , 2018, GECCO.

[37]  Peter A. N. Bosman,et al.  Convolutional neural network surrogate-assisted GOMEA , 2019, GECCO.

[38]  Tanja Alderliesten,et al.  GPU-Accelerated Bi-Objective Treatment Planning for Prostate High-Dose-Rate Brachytherapy. , 2019, Medical physics.

[39]  Marco Virgolin,et al.  Delft University of Technology Machine learning for the prediction of pseudorealistic pediatric abdominal phantoms for radiation dose reconstruction , 2020 .

[40]  Tanja Alderliesten,et al.  Leveraging conditional linkage models in gray-box optimization with the real-valued gene-pool optimal mixing evolutionary algorithm , 2020, GECCO.

[41]  Michal Przewozniczek,et al.  Empirical Linkage Learning , 2020, IEEE Transactions on Evolutionary Computation.

[42]  E. Goodman Human-competitive results awards - "Humies" 2019 - announces winners at GECCO , 2020 .

[43]  Yusuke Nojima,et al.  Towards realistic optimization benchmarks: a questionnaire on the properties of real-world problems , 2020, GECCO Companion.

[44]  Peter A. N. Bosman,et al.  A Novel Approach to Designing Surrogate-assisted Genetic Algorithms by Combining Efficient Learning of Walsh Coefficients and Dependencies , 2021, ACM Trans. Evol. Learn. Optim..