micro-population Evolutionary algorithms (mu-EAs) are useful tools for optimization purposes. They can be used as optimizers for unconstrained, constraint and multi-objective problems. mu-EAs distinctive feature is th...
详细信息
ISBN:
(纸本)9780769539331
micro-population Evolutionary algorithms (mu-EAs) are useful tools for optimization purposes. They can be used as optimizers for unconstrained, constraint and multi-objective problems. mu-EAs distinctive feature is the usage of very small populations. A novel mu-EA named Elitistic Evolution (EEv) is proposed in paper. EEv is designed to solve high-dimensionality problems (N >= 30) without using complex mechanisms e.g. Hessian or covariance matrix. It is a simple heuristic that does not require a careful fine-tunning of its parameters. EEv principal features are: adaptive behavior and elitism. Its evolutionary operators: mutation, crossover and replacement, have the ability to search either locally (near a current point) or globally (on a distant point). This ability is controlled by a single adaptive parameter. EEv is tested on a set of well-known optimization problems and its performance is compared with respect to state-of-the-art algorithms, such as Differential Evolution, mu-PSO and Restart CMA-ES.
The loser-out tournament-based firework algorithm (LoTFWA) is a new baseline among firework algorithm (FWA) variants due to its outstanding performance in multimodal optimization problems. LoTFWA successfully achieves...
详细信息
ISBN:
(纸本)9783031096778;9783031096761
The loser-out tournament-based firework algorithm (LoTFWA) is a new baseline among firework algorithm (FWA) variants due to its outstanding performance in multimodal optimization problems. LoTFWA successfully achieves information-interaction among populations by introducing a competition mechanism, while information-interaction within each sub-population remains insufficient. To solve this issue, this paper proposes a micro-population evolution strategy and a hybrid algorithm LoTFWA-microDE. Under the proposed strategy, sparks generated by one firework make up a micro-population which is taken into the differential evolution procedure. The proposed algorithm is tested on the CEC' 13 benchmark functions. Experimental results show that the proposed algorithm attains significantly better performance than LoTFWA and DE in multi-modal functions, which indicates the superiority of the proposed micro-population evolution strategy.
暂无评论