Multimodal multi-objective optimization: A preliminary study

In real world applications, there are many multi-objective optimization problems. Most existing multi-objective optimization algorithms focus on improving the diversity, spread and convergence of the solutions in the objective space. Few works study the distribution of solutions in the decision space. In practical applications, some multi-objective problems have different Pareto sets with the same objective values and these problems are defined as multimodal multi-objective optimization problems. It is of great significance to provide all the Pareto sets for the decision maker. This paper describes the concept of multimodal multi-objective optimization problems in detail. Novel test functions are also designed to judge the performance of different algorithms. Moreover, some existing multi-objective algorithms are tested and compared. Finally, a decision space based niching multi-objective evolutionary algorithm is proposed to solve these problems. The experimental results suggest that existing multi-objective optimization algorithms fail to find all the Pareto sets while the proposed algorithm is able to find almost all the Pareto sets without deteriorating the distribution of solutions in the objective space.

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

[2]  Marco Laumanns,et al.  SPEA2: Improving the strength pareto evolutionary algorithm , 2001 .

[3]  Qingfu Zhang,et al.  Multiobjective optimization Test Instances for the CEC 2009 Special Session and Competition , 2009 .

[4]  A. Hamzeh,et al.  A Multi-Objective Evolutionary Algorithm based on complete-linkage clustering to enhance the solution space diversity , 2012, The 16th CSI International Symposium on Artificial Intelligence and Signal Processing (AISP 2012).

[5]  Ofer M. Shir,et al.  Enhancing Decision Space Diversity in Evolutionary Multiobjective Algorithms , 2009, EMO.

[6]  Kalyanmoy Deb,et al.  Multi-Objective Evolutionary Algorithms , 2015, Handbook of Computational Intelligence.

[7]  Kalyanmoy Deb,et al.  Omni-optimizer: A Procedure for Single and Multi-objective Optimization , 2005, EMO.

[8]  David W. Coit,et al.  Multi-objective optimization using genetic algorithms: A tutorial , 2006, Reliab. Eng. Syst. Saf..

[9]  Nikolaus Hansen,et al.  Completely Derandomized Self-Adaptation in Evolution Strategies , 2001, Evolutionary Computation.

[10]  Eckart Zitzler,et al.  Integrating decision space diversity into hypervolume-based multiobjective search , 2010, GECCO '10.

[11]  Erkan Ülker,et al.  B-Spline Curve Knot Estimation by Using Niched Pareto Genetic Algorithm (NPGA) , 2016 .

[12]  Xiaodong Li,et al.  Niching Without Niching Parameters: Particle Swarm Optimization Using a Ring Topology , 2010, IEEE Transactions on Evolutionary Computation.

[13]  Qingfu Zhang,et al.  MOEA/D: A Multiobjective Evolutionary Algorithm Based on Decomposition , 2007, IEEE Transactions on Evolutionary Computation.

[14]  Qingfu Zhang,et al.  Approximating the Set of Pareto-Optimal Solutions in Both the Decision and Objective Spaces by an Estimation of Distribution Algorithm , 2009, IEEE Transactions on Evolutionary Computation.

[15]  Mike Preuss,et al.  Multimodal Optimization by Means of Evolutionary Algorithms , 2015, Natural Computing Series.

[16]  P. John Clarkson,et al.  A Species Conserving Genetic Algorithm for Multimodal Function Optimization , 2002, Evolutionary Computation.

[17]  Jing J. Liang,et al.  Differential Evolution With Neighborhood Mutation for Multimodal Optimization , 2012, IEEE Transactions on Evolutionary Computation.

[18]  D. J. Cavicchio,et al.  Adaptive search using simulated evolution , 1970 .

[19]  David E. Goldberg,et al.  A niched Pareto genetic algorithm for multiobjective optimization , 1994, Proceedings of the First IEEE Conference on Evolutionary Computation. IEEE World Congress on Computational Intelligence.

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

[21]  Michael T. M. Emmerich,et al.  Test Problems Based on Lamé Superspheres , 2007, EMO.

[22]  Mike Preuss,et al.  Niching Methods and Multimodal Optimization Performance , 2015 .

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

[24]  K. Dejong,et al.  An analysis of the behavior of a class of genetic adaptive systems , 1975 .

[25]  Günter Rudolph,et al.  Pareto Set and EMOA Behavior for Simple Multimodal Multiobjective Functions , 2006, PPSN.

[26]  Alain Pétrowski,et al.  A clearing procedure as a niching method for genetic algorithms , 1996, Proceedings of IEEE International Conference on Evolutionary Computation.