A new selection mechanism based on hypervolume and its locality property

In this paper, we propose a new selection mechanism based on the hypervolume indicator and on its “locality property”, which is incorporated into the SMSEMOA, giving rise to the so-called improved SMS-EMOA (iSMS-EMOA). Our proposed selection mechanism is validated using standard test functions taken from the specialized literature, having three to six objective functions. iSMS-EMOA is compared with respect to its predecessor SMS-EMOA and with respect to another version of SMS-EMOA that uses the approximation of the hypervolume indicator, instead of its exact calculation. Our preliminary results indicate that our proposed selection mechanism outperforms the selection mechanisms based on the hypervolume indicator that have been proposed in recent years, since it significantly reduces the computational time required by the algorithm without sacrificing quality in the approximation generated.

[1]  Lucas Bradstreet,et al.  Updating exclusive hypervolume contributions cheaply , 2009, 2009 IEEE Congress on Evolutionary Computation.

[2]  Hisao Ishibuchi,et al.  Indicator-based evolutionary algorithm with hypervolume approximation by achievement scalarizing functions , 2010, GECCO '10.

[3]  Nicola Beume,et al.  SMS-EMOA: Multiobjective selection based on dominated hypervolume , 2007, Eur. J. Oper. Res..

[4]  Marco Laumanns,et al.  Scalable Test Problems for Evolutionary Multiobjective Optimization , 2005, Evolutionary Multiobjective Optimization.

[5]  Anne Auger,et al.  Theoretically Investigating Optimal µ-Distributions for the Hypervolume Indicator: First Results for Three Objectives , 2010, PPSN.

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

[7]  Anne Auger,et al.  Theory of the hypervolume indicator: optimal μ-distributions and the choice of the reference point , 2009, FOGA '09.

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

[9]  Tobias Friedrich,et al.  Convergence of hypervolume-based archiving algorithms ii: competitiveness , 2012, GECCO.

[10]  Carlos M. Fonseca,et al.  Computing Hypervolume Contributions in Low Dimensions: Asymptotically Optimal Algorithm and Complexity Results , 2011, EMO.

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

[12]  Eckart Zitzler,et al.  Improving hypervolume-based multiobjective evolutionary algorithms by using objective reduction methods , 2007, 2007 IEEE Congress on Evolutionary Computation.

[13]  Lothar Thiele,et al.  Multiobjective Optimization Using Evolutionary Algorithms - A Comparative Case Study , 1998, PPSN.

[14]  Stefan Roth,et al.  Covariance Matrix Adaptation for Multi-objective Optimization , 2007, Evolutionary Computation.

[15]  David W. Corne,et al.  Properties of an adaptive archiving algorithm for storing nondominated vectors , 2003, IEEE Trans. Evol. Comput..

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

[17]  M. Farina,et al.  On the optimal solution definition for many-criteria optimization problems , 2002, 2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622).

[18]  Jürgen Branke,et al.  Multi-objective particle swarm optimization on computer grids , 2007, GECCO '07.

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

[20]  Eckart Zitzler,et al.  HypE: An Algorithm for Fast Hypervolume-Based Many-Objective Optimization , 2011, Evolutionary Computation.