Title :
Enhancing the robustness of a speciation-based PSO
Author :
Bird, Stefan ; Li, Xiaodong
Author_Institution :
RMIT Univ., Melbourne
Abstract :
Speciation encourages an evolutionary algorithm to locate multiple solutions in multimodal environments. Speciation algorithms often require a user to specify a parameter to define the species radius, which can be a major drawback since this knowledge may not be available a priori. This paper proposes a technique using a time-based convergence measure to overcome this problem. The proposed method is used to enhance the performance of a speciation-based PSO (SPSO) and has been shown to be robust over a wide range of values for this user-specified parameter.
Keywords :
evolutionary computation; particle swarm optimisation; evolutionary algorithm; speciation-based PSO; time-based convergence; user-specified parameter; Birds; Computer science; Convergence; Evolutionary computation; Information technology; Particle swarm optimization; Robustness; Space exploration; Stochastic processes; Testing;
Conference_Titel :
Evolutionary Computation, 2006. CEC 2006. IEEE Congress on
Conference_Location :
Vancouver, BC
Print_ISBN :
0-7803-9487-9
DOI :
10.1109/CEC.2006.1688399