A first mathematical runtime analysis of the non-dominated sorting genetic algorithm II (NSGA-II): (hot-off-the-press track at GECCO 2022)

The non-dominated sorting genetic algorithm II (NSGA-II) is the most intensively used multi-objective evolutionary algorithm (MOEA) in real-world applications. However, in contrast to several simple MOEAs analyzed also via mathematical means, no such study exists for the NSGA-II so far. In this work, we show that mathematical runtime analyses are feasible also for the NSGA-II. As particular results, we prove that with a population size larger than the Pareto front size by a constant factor, the NSGA-II with two classic mutation operators and three different ways to select the parents satisfies the same asymptotic runtime guarantees as the SEMO and GSEMO algorithms on the basic ONEMINMAX and LOTZ benchmark functions. However, if the population size is only equal to the size of the Pareto front, then the NSGA-II cannot efficiently compute the full Pareto front (for an exponential number of iterations, the population will always miss a constant fraction of the Pareto front). Our experiments confirm

[1]  Frank Neumann,et al.  Pareto Optimization for Subset Selection with Dynamic Cost Constraints , 2018, AAAI.

[2]  Benjamin Doerr,et al.  Theoretical Analyses of Multi-Objective Evolutionary Algorithms on Multi-Modal Objectives , 2020, ArXiv.

[3]  Chao Qian,et al.  A General Approach to Running Time Analysis of Multi-objective Evolutionary Algorithms , 2018, IJCAI.

[4]  Yuren Zhou,et al.  Running Time Analysis of MOEA/D with Crossover on Discrete Optimization Problem , 2019, AAAI.

[5]  Frank Neumann,et al.  Population size matters: rigorous runtime results for maximizing the hypervolume indicator , 2013, GECCO '13.

[6]  Frank Neumann,et al.  Analyzing Hypervolume Indicator Based Algorithms , 2008, PPSN.

[7]  Yuren Zhou,et al.  A Primary Theoretical Study on Decomposition-Based Multiobjective Evolutionary Algorithms , 2016, IEEE Transactions on Evolutionary Computation.

[8]  Yang Yu,et al.  An Efficient Evolutionary Algorithm for Subset Selection with General Cost Constraints , 2020, AAAI.

[9]  Qingfu Zhang,et al.  Multiobjective evolutionary algorithms: A survey of the state of the art , 2011, Swarm Evol. Comput..

[10]  Marco Laumanns,et al.  Running time analysis of a multi-objective evolutionary algorithm on a simple discrete optimization problem , 2002 .

[11]  Victoria G. Crawford.  Faster Guarantees of Evolutionary Algorithms for Maximization of Monotone Submodular Functions , 2019, IJCAI.

[12]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[13]  Yang Yu,et al.  On Subset Selection with General Cost Constraints , 2017, IJCAI.

[14]  Zhengxin Huang,et al.  Runtime Analysis of Somatic Contiguous Hypermutation Operators in MOEA/D Framework , 2020, AAAI.

[15]  Frank Neumann,et al.  Maximizing Submodular Functions under Matroid Constraints by Evolutionary Algorithms , 2015, Evolutionary Computation.

[16]  Marco Laumanns,et al.  Running time analysis of multiobjective evolutionary algorithms on pseudo-Boolean functions , 2004, IEEE Transactions on Evolutionary Computation.

[17]  Carsten Witt,et al.  Approximating Covering Problems by Randomized Search Heuristics Using Multi-Objective Models , 2007, Evolutionary Computation.

[18]  Dirk Sudholt,et al.  Design and Analysis of Diversity-Based Parent Selection Schemes for Speeding Up Evolutionary Multi-objective Optimisation , 2018, Theor. Comput. Sci..

[19]  Frank Neumann,et al.  Computing Minimum Cuts by Randomized Search Heuristics , 2008, GECCO '08.

[20]  Benjamin Doerr,et al.  Lower bounds for the runtime of a global multi-objective evolutionary algorithm , 2013, 2013 IEEE Congress on Evolutionary Computation.

[21]  Zhi-Hua Zhou,et al.  Maximizing submodular or monotone approximately submodular functions by multi-objective evolutionary algorithms , 2017, Artif. Intell..

[22]  Chao Bian,et al.  Subset Selection by Pareto Optimization with Recombination , 2020, AAAI.

[23]  Oliver Giel,et al.  Expected runtimes of a simple multi-objective evolutionary algorithm , 2003, The 2003 Congress on Evolutionary Computation, 2003. CEC '03..

[24]  Yang Yu,et al.  An analysis on recombination in multi-objective evolutionary optimization , 2011, GECCO '11.