Adaptive Differential Evolution With Sorting Crossover Rate for Continuous Optimization Problems

• Published in: IEEE transactions on cybernetics Vol. 47; no. 9; pp. 2742 - 2753
• Main Authors: Zhou, Yin-Zhi, Yi, Wen-Chao, Gao, Liang, Li, Xin-Yu
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.09.2017; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Adaptive differential evolution (JADE); Algorithm design and analysis; Algorithms; Classification; Convergence; Evolutionary algorithms; Fitness; Gaussian distribution; Manufacturing; Normal distribution; Optimization; Performance enhancement; scheme retention mechanism; Sociology; Sorting; Sorting algorithms; sorting crossover rate (<italic xmlns:ali="http://www.niso.org/schemas/ali/1.0/" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">CR ); Statistics
• ISSN: 2168-2267; 2168-2275; 2168-2275
• DOI: 10.1109/TCYB.2017.2676882

Differential evolution (DE) is one of the best evolutionary algorithms (EAs). The effort of improving its performance has received great research attentions, such as adaptive DE (JADE). Based on the analysis on the aspects that may improve the performance of JADE, we introduce a modified JADE version with sorting crossover rate (CR). In JADE, CR values are generated based on mean value and Gaussian distribution. In the proposed algorithm, a smaller CR value is assigned to individual with better fitness value. Therefore, the components of the individuals, which have better fitness values, can appear in the offspring with higher possibility. In addition, the better offspring generated from last iteration are supposed to have better schemes, hence these schemes are preserved in next offspring generation procedure. This modified version is called as JADE algorithm with sorting CR (JADE_sort). The experiments results with several excellent algorithms show the effectiveness of JADE_sort.