Abstract | ||
---|---|---|
Both Particle Swarm Optimization (PSO) and Evolution Strategy with Covariance Matrix Adaptation (CMA-ES) exhibit good performance when solving global optimization problems. However, PSO could be misled by historical information and falls into a local optimum. Further, CMA-ES cannot fully utilize global information. Therefore, in this paper, we first propose a time-window PSO (TW-PSO) as an improvement of PSO, which could enhance the exploration ability of the algorithm. Second, we design a hybrid algorithm of TW-PSO, PSO and CMA-ES, i.e., HTPC, which combines the advantages of TW-PSO, PSO, and CMA-ES. We test HTPC on single-objective optimization problems from the CEC-2019 100-Digit Challenge, and the experimental results show that the performance of HTPC is competitive. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1109/CEC.2019.8789912 | 2019 IEEE Congress on Evolutionary Computation (CEC) |
Keywords | Field | DocType |
Global Optimization,Particle Swarm Optimization,CMA-ES | Particle swarm optimization,Mathematical optimization,Hybrid algorithm,Global optimization,Computer science,Local optimum,Network topology,Evolution strategy,CMA-ES,Optimization problem | Conference |
ISBN | Citations | PageRank |
978-1-7281-2154-3 | 0 | 0.34 |
References | Authors | |
21 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Peilan Xu | 1 | 1 | 1.70 |
Wenjian Luo | 2 | 356 | 40.95 |
Xin Lin | 3 | 26 | 12.01 |
Yingying Qiao | 4 | 1 | 2.71 |
Tao Zhu | 5 | 0 | 0.34 |