A fast many-objective hypervolume algorithm using iterated incremental calculations

Three fast algorithms have been proposed for calculating hypervolume exactly: the Hypervolume by Slicing Objectives algorithm (HSO) optimised with heuristics designed to improve the average case; an adaptation of the Overmars and Yap algorithm for solving the Klee's measure problem; and a recent algorithm by Fonseca et al. We propose a fourth algorithm IIHSO based largely on the Incremental HSO algorithm, a version of HSO adapted to calculate the exclusive hypervolume contribution of a point to a front. We give a comprehensive analysis of IIHSO and performance comparison between three state of the art algorithms, and conclude that IIHSO outperforms the others on most important and representative data in many objectives.

[1]  Joshua D. Knowles.  Local-search and hybrid evolutionary algorithms for Pareto optimization , 2002 .

[2]  Nicola Beume,et al.  Pareto-, Aggregation-, and Indicator-Based Methods in Many-Objective Optimization , 2007, EMO.

[3]  R. Lyndon While,et al.  A review of multiobjective test problems and a scalable test problem toolkit , 2006, IEEE Transactions on Evolutionary Computation.

[4]  Lucas Bradstreet,et al.  A Fast Incremental Hypervolume Algorithm , 2008, IEEE Transactions on Evolutionary Computation.

[5]  Nicola Beume,et al.  An EMO Algorithm Using the Hypervolume Measure as Selection Criterion , 2005, EMO.

[6]  Mark Fleischer,et al.  The measure of pareto optima: Applications to multi-objective metaheuristics , 2003 .

[7]  Marco Laumanns,et al.  Performance assessment of multiobjective optimizers: an analysis and review , 2003, IEEE Trans. Evol. Comput..

[8]  R. Lyndon While,et al.  A faster algorithm for calculating hypervolume , 2006, IEEE Transactions on Evolutionary Computation.

[9]  Nicola Beume,et al.  S-Metric Calculation by Considering Dominated Hypervolume as Klee's Measure Problem , 2009, Evolutionary Computation.

[10]  Joshua D. Knowles,et al.  Bounded archiving using the lebesgue measure , 2003, The 2003 Congress on Evolutionary Computation, 2003. CEC '03..

[11]  Kalyanmoy Deb,et al.  Faster Hypervolume-Based Search Using Monte Carlo Sampling , 2008, MCDM.

[12]  Mark H. Overmars,et al.  New upper bounds in Klee's measure problem , 1988, [Proceedings 1988] 29th Annual Symposium on Foundations of Computer Science.

[13]  Jonathan E. Fieldsend,et al.  Full Elite Sets for Multi-Objective Optimisation , 2002 .

[14]  Eckart Zitzler,et al.  Indicator-Based Selection in Multiobjective Search , 2004, PPSN.

[15]  Luigi Barone,et al.  An evolution strategy with probabilistic mutation for multi-objective optimisation , 2003, The 2003 Congress on Evolutionary Computation, 2003. CEC '03..

[16]  Zbigniew Michalewicz,et al.  Handbook of Evolutionary Computation , 1997 .

[17]  Lucas Bradstreet,et al.  Heuristics for optimizing the calculation of hypervolume for multi-objective optimization problems , 2005, 2005 IEEE Congress on Evolutionary Computation.

[18]  Marco Laumanns,et al.  A unified model for multi-objective evolutionary algorithms with elitism , 2000, Proceedings of the 2000 Congress on Evolutionary Computation. CEC00 (Cat. No.00TH8512).

[19]  Marco Laumanns,et al.  Scalable multi-objective optimization test problems , 2002, Proceedings of the 2002 Congress on Evolutionary Computation. CEC'02 (Cat. No.02TH8600).

[20]  Eckart Zitzler,et al.  Evolutionary algorithms for multiobjective optimization: methods and applications , 1999 .

[21]  Carlos M. Fonseca,et al.  An Improved Dimension-Sweep Algorithm for the Hypervolume Indicator , 2006, 2006 IEEE International Conference on Evolutionary Computation.

[22]  Nicola Beume,et al.  On the Complexity of Computing the Hypervolume Indicator , 2009, IEEE Transactions on Evolutionary Computation.

[23]  Robin Charles Purshouse,et al.  On the evolutionary optimisation of many objectives , 2003 .

[24]  Luís Paquete,et al.  An optimal algorithm for a special case of Klee's measure problem in three dimensions , 2006 .