Many-objective Evolutionary Optimization by Self-Controlling Dominance Area of Solutions :
-
- Sato Hiroyuki
- Faculty of Informatics and Engineering, The University of Electro-Communications
-
- Aguirre Hernán E.
- International Young Researcher Empowerment Center, Shinshu University
-
- Tanaka Kiyoshi
- Faculty of Engineering, Shinshu University
Bibliographic Information
- Other Title
-
- 解の支配領域の自己制御による進化型多数目的最適化:
- 多数目的 0/1 ナップザック問題における性能検証と挙動解析
- Performance Verification and Analysis in Many-objective 0/1 Knapsack Problems
Abstract
Controlling dominance area of solutions (CDAS) relaxes the concepts of Pareto dominance with an user defined parameter S. This method enhances the search performance of dominance-based MOEA in many-objective optimization problems (MaOPs). However, to bring out desirable search performance, we have to experimentally find out S that controls dominance areas appropriately. Also, there is a tendency to deteriorate the diversity of solutions obtained by CDAS when we decrease S from 0.5. To solve these problems, in this work, we propose a modification of CDAS called self-controlling dominance area of solutions (S-CDAS). In S-CDAS, the algorithm self-controls dominance areas for each solution without the need of an external parameter. S-CDAS considers convergence and diversity and realizes a fine grained ranking that is different from conventional CDAS. In this work, we focus on combinatorial optimization and use many-objective 0/1 knapsack problems with m = 4∼10 objectives to verify the search performance of the proposed method. Simulation results show that S-CDAS achieves well-balanced search performance on both convergence and diversity compared to conventional NSGA-II, CDAS, IBEAε+ and MSOPS. In addition, the algorithm's behavior of S-CDAS is analyzed and discussed.
Journal
-
- Transaction of the Japanese Society for Evolutionary Computation
-
Transaction of the Japanese Society for Evolutionary Computation 1 (1), 32-42, 2010
The Japanese Society for Evolutionary Computation
- Tweet
Keywords
Details 詳細情報について
-
- CRID
- 1390282680342249984
-
- NII Article ID
- 130004965133
-
- ISSN
- 21857385
-
- Text Lang
- ja
-
- Data Source
-
- JaLC
- CiNii Articles
-
- Abstract License Flag
- Disallowed