Adaptive evolutionary programming with p-best mutation strategy

被引:9
作者
Das, Swagatam [1 ]
Mallipeddi, Rammohan [3 ]
Maity, Dipankar [2 ]
机构
[1] Indian Stat Inst, Elect & Commun Sci Unit, Kolkata 700108, India
[2] Jadavpur Univ, Dept Elect & Telecommun Engn, Kolkata 700032, W Bengal, India
[3] Nanyang Technol Univ, Sch Elect & Elect Engn, Singapore 639798, Singapore
关键词
Evolutionary programming; Real-parameter optimization; Mutation; Parent selection; Evolutionary algorithms; DIFFERENTIAL EVOLUTION; GLOBAL OPTIMIZATION; PARAMETERS;
D O I
10.1016/j.swevo.2012.11.002
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Although initially conceived for evolving finite state machines, Evolutionary Programming (EP), in its present form, is largely used as a powerful real parameter optimizer. For function optimization. EP mainly relies on its mutation operators. Over past few years several mutation operators have been proposed to improve the performance of EP on a wide variety of numerical benchmarks. However, unlike real-coded GAs, there has been no fitness-induced bias in parent selection for mutation in EP. That means the i-th population member is selected deterministically for mutation and creation of the i-th offspring in each generation. In this article we propose a p-best mutation scheme for EP where any one from the p (p is an element of [1,2, ... ,mu], where mu denotes population size) top-ranked population-members (according to fitness values) is selected randomly for mutation. The scheme is invoked with 50% probability with each index in the current population, i.e. the i-th offspring can now be obtained either by mutating the i-th parent or by mutating a randomly selected individual from the p top-ranked vectors. The percentage of best members is made dynamic by decreasing p in from mu/2 to 1 with generations to favor explorative behavior at the early stages of search and exploitation during the later stages. We investigate the effectiveness of introducing controlled bias in parent selection in conjunction with an Adaptive Fast EP (AFEP), where the value of a strategy parameter is updated based on the previous records of successful mutations by the same parameter. Comparison with the recent and best-known versions of EP over 25 benchmark functions from the CEC (Congress on Evolutionary Computation) 2005 test-suite for real-parameter optimization and two other engineering optimization problems reflects the statistically validated superiority of the new scheme in terms of final accuracy, speed, and robustness. Comparison with AFEP without p-best mutation demonstrates the improvement of performance due to the proposed mutation scheme alone. (C) 2012 Elsevier B.V. All rights reserved.
引用
收藏
页码:58 / 68
页数:11
相关论文
共 33 条
[1]  
[Anonymous], THESIS
[2]  
[Anonymous], 1966, Artificial_Intelligence_Through_Simulated Evolution
[3]  
Augur A., 2009, P IEEE C EV COMP
[4]   Self-adapting control parameters in differential evolution: A comparative study on numerical benchmark problems [J].
Brest, Janez ;
Greiner, Saso ;
Boskovic, Borko ;
Mernik, Marjan ;
Zumer, Vijern .
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2006, 10 (06) :646-657
[5]  
Chellapilla K., 1998, IEEE Transactions on Evolutionary Computation, V2, P91, DOI 10.1109/4235.735431
[6]   Preserving and Exploiting Genetic Diversity in Evolutionary Programming Algorithms [J].
Chen, Gang ;
Low, Chor Ping ;
Yang, Zhonghua .
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2009, 13 (03) :661-673
[7]   Differential Evolution Using a Neighborhood-Based Mutation Operator [J].
Das, Swagatam ;
Abraham, Ajith ;
Chakraborty, Uday K. ;
Konar, Amit .
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2009, 13 (03) :526-553
[8]   Evolutionary programming using a mixed mutation strategy [J].
Dong, Hongbin ;
He, Jun ;
Huang, Houkuan ;
Hou, Wei .
INFORMATION SCIENCES, 2007, 177 (01) :312-327
[9]   A METHOD OF A SPREAD-SPECTRUM RADAR POLYPHASE CODE DESIGN [J].
DUKIC, ML ;
DOBROSAVLJEVIC, ZS .
IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, 1990, 8 (05) :743-749
[10]  
Eiben A.E., 2015, NAT COMP SER