Control parameter sensitivity analysis of the multi-guide particle swarm optimization algorithm

This paper conducts a sensitivity analysis of the recently proposed multi-objective optimizing algorithm, namely the multi-guide particle swarm optimization algorithm (MGPSO). The MGPSO uses subswarms to explore the search space, where each subswarm optimises one of the multiple objectives. A bounded archive is used to share previously found non-dominated solutions between subswarms. A third term, the archive guide, is added to the velocity update equation that represents a randomly selected solution from the archive. The influence of the archive guide on a particle is controlled by the archive balance coefficient and is proportional to the social guide. The original implementation of the MGPSO used static values randomly sampled from a uniform distribution in the range [0,1] for the archive balance coefficient. This paper investigates a number of approaches to dynamically adjust this control parameter. These approaches are evaluated on a variety of multi-objective optimization problems. It is shown that a linearly increasing strategy and stochastic strategies outperformed the standard approach to initializing the archive balance coefficient on two-objective and three objective optimization problems.

[1]  Antonio J. Nebro,et al.  Redesigning the jMetal Multi-Objective Optimization Framework , 2015, GECCO.

[2]  L. Miles,et al.  2000 , 2000, RDH.

[3]  Gary B. Lamont,et al.  Multiobjective evolutionary algorithm test suites , 1999, SAC '99.

[4]  S. M. García,et al.  2014: , 2020, A Party for Lazarus.

[5]  Carlos M. Fonseca,et al.  Evolutionary Multi-Criterion Optimization , 2019, Lecture Notes in Computer Science.

[6]  D.A. Van Veldhuizen,et al.  On measuring multiobjective evolutionary algorithm performance , 2000, Proceedings of the 2000 Congress on Evolutionary Computation. CEC00 (Cat. No.00TH8512).

[7]  Andries Petrus Engelbrecht,et al.  Particle swarm convergence: An empirical investigation , 2014, 2014 IEEE Congress on Evolutionary Computation (CEC).

[8]  Martin J. Oates,et al.  PESA-II: region-based selection in evolutionary multiobjective optimization , 2001 .

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

[10]  Riccardo Poli,et al.  Particle swarm optimization , 1995, Swarm Intelligence.

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

[12]  Russell C. Eberhart,et al.  A new optimizer using particle swarm theory , 1995, MHS'95. Proceedings of the Sixth International Symposium on Micro Machine and Human Science.

[13]  Yue Shi,et al.  A modified particle swarm optimizer , 1998, 1998 IEEE International Conference on Evolutionary Computation Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98TH8360).

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

[15]  J. van Leeuwen,et al.  Evolutionary Multi-Criterion Optimization , 2003, Lecture Notes in Computer Science.

[16]  Kalyanmoy Deb,et al.  A Fast Elitist Non-dominated Sorting Genetic Algorithm for Multi-objective Optimisation: NSGA-II , 2000, PPSN.

[17]  Peter J. Fleming,et al.  On the Performance Assessment and Comparison of Stochastic Multiobjective Optimizers , 1996, PPSN.

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

[19]  John Fulcher,et al.  Computational Intelligence: An Introduction , 2008, Computational Intelligence: A Compendium.

[20]  Carlos A. Coello Coello,et al.  Applications of multi-objective evolutionary algorithms in economics and finance: A survey , 2007, 2007 IEEE Congress on Evolutionary Computation.