IEEE Trans Cybern. 2015 Sep;45(9):1798-810. doi: 10.1109/TCYB.2014.2360752. Epub 2014 Oct 9.
Utilizing cumulative correlation information already existing in an evolutionary process, this paper proposes a predictive approach to the reproduction mechanism of new individuals for differential evolution (DE) algorithms. DE uses a distributed model (DM) to generate new individuals, which is relatively explorative, whilst evolution strategy (ES) uses a centralized model (CM) to generate offspring, which through adaptation retains a convergence momentum. This paper adopts a key feature in the CM of a covariance matrix adaptation ES, the cumulatively learned evolution path (EP), to formulate a new evolutionary algorithm (EA) framework, termed DEEP, standing for DE with an EP. Without mechanistically combining two CM and DM based algorithms together, the DEEP framework offers advantages of both a DM and a CM and hence substantially enhances performance. Under this architecture, a self-adaptation mechanism can be built inherently in a DEEP algorithm, easing the task of predetermining algorithm control parameters. Two DEEP variants are developed and illustrated in the paper. Experiments on the CEC'13 test suites and two practical problems demonstrate that the DEEP algorithms offer promising results, compared with the original DEs and other relevant state-of-the-art EAs.
本文利用进化过程中已经存在的累积相关信息，提出了一种针对差分进化（DE）算法的新个体繁殖机制的预测方法。DE 使用分布式模型（DM）生成新个体，相对具有探索性，而进化策略（ES）使用集中式模型（CM）生成后代，通过适应保留收敛动力。本文采用协方差矩阵自适应 ES 的 CM 中的一个关键特征，即累积学习的进化路径（EP），来构建一种新的进化算法（EA）框架，称为 DEEP，代表具有 EP 的 DE。无需将两种基于 CM 和 DM 的算法机械地结合在一起，DEEP 框架具有 DM 和 CM 的优势，从而大大提高了性能。在这种架构下，DEEP 算法可以内在地构建自适应机制，减轻预先确定算法控制参数的任务。本文还开发并说明了两种 DEEP 变体。CEC'13 测试套件和两个实际问题的实验表明，与原始 DE 和其他相关最先进的 EA 相比，DEEP 算法提供了有希望的结果。
