Social interaction in particle swarm optimization, the ranked FIPS, and adaptive multi-swarms

ABSTRACT

The interaction among particles is a vital aspect of Particle Swarm Optimization. As such, it has a strong influence on the swarm's success. In this study various approaches regarding the particles' communication behavior and their relationship are examined, as well as possibilities to combine the approaches. A new variant of the popular FIPS algorithm, the so-called Ranked FIPS, is introduced, which resolves specific shortcomings of the traditional FIPS. As all tested PSO variants feature distinct strengths and weaknesses, a new adaptive strategy is proposed which operates on dissimiliarly configured subswarms. The exchange between these subswarms is solely based on particle migration. The combination of the Ranked FIPS and other strategies within the so called Particle Swarm Optimizer with Migration achieves a very good, yet remarkably reliable performance over a wide range of recognized benchmark problems.

REFERENCES

• 1: James Kennedy , Russell C. Eberhart, Swarm intelligence, Morgan Kaufmann Publishers Inc., San Francisco, CA, 2001
• 2: D. Bratton and J. Kennedy. Defining a standard for Particle Swarm Optimization. In Proceedings of the 2007 IEEE Swarm Intelligence Symposium, 2007.
• 3: J. Kennedy. Stereotyping: Improving particle swarm performance with cluster analysis. In Proceedings of the 2000 Congress on Evolutionary Computation, pages 150--151. IEEE Service Center, 2000.
• 4: R. Mendes, J. Kennedy, and J. Neves. The fully informed particle swarm: Simpler, maybe better. IEEE Transactions on Evolutionary Computation, 8(3):204--210, 2004.
• 5: J. Kennedy and R. Mendes. Neighborhood topologies in fully informed and best-of-neighborhood particle swarms. IEEE Transactions on Systems, Man and Cybernetics, 36(4):515--519, July 2006.
• 6: P. N. Suganthan. Particle swarm optimiser with neighbourhood operator. In Proceedings of the Congress on Evolutionary Computation, volume 3, pages 1958--1962. IEEE Press, 6-9 July 1999.
• 7: M. Richards and D. Ventura. Dynamic sociometry in particle swarm optimization. In Proceedings of the Sixth International Conference on Computational Intelligence and Natural Computing, pages 1557--1560, North Carolina, September 2003.
• 8: R. Mendes, J. Kennedy, and J. Neves. Watch thy neighbor or how the swarm can learn from its environment. In Proceedings of the 2003 IEEE Swarm Intelligence Symposium, pages 88--94. IEEE, 2003.
• 9: P. N. Suganthan, N. Hansen, J. J. Liang, K. Deb, Y. P. Chen, A. Auger, and S. Tiwari. Problem Definitions and Evaluation Criteria for the CEC 2005 Special Session on Real--Parameter Optimization. Technical report, Nanyang Technological University, Singapore, 2005.
• 10: J. Kennedy , R. Mendes, Population structure and particle swarm performance, Proceedings of the Evolutionary Computation on 2002. CEC '02. Proceedings of the 2002 Congress, p.1671-1676, May 12-17, 2002

INDEX TERMS

Primary Classification:
  G. Mathematics of Computing
  G.1 NUMERICAL ANALYSIS
      G.1.6 Optimization
          Subjects: Global optimization

Additional Classification:
  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE
      I.2.8 Problem Solving, Control Methods, and Search
          Subjects: Heuristic methods

General Terms:
Algorithms

Keywords:
adaptive optimization, particle swarm optimization, performance analysis, social interaction, subswarms