Using Opposition-based Learning with Particle Swarm Optimization and Barebones Differential Evolution
Mahamed G. H. Omran · InTech eBooks · 2009
Opposition-based learning was used in this chapter to improve the performance of PSO and BBDE. Two opposition-based variants were proposed (namely, iPSO and iBBDE). The iPSO and iBBDE algorithms replace the least-fit particle with its anti-particle. The results show that, in general, iPSO and iBBDE outperformed PSO and BBDE, respectively. In addition, the results show that using OBL enhances the performance of PSO and BBDE without requiring additional parameters. The ideas introduced in this chapter could also be used with any PSO/BBDE variant. Future research will investigate the effect of noise on the performance of the proposed approaches. Furthermore, a scalability study will be conducted. Finally, applying the proposed approaches to real-world problem will be investigated.