Title :
Active covariance matrix adaptation for multi-objective CMA-ES
Author :
Krimpmann, Christoph ; Braun, Johannes ; Hoffmann, F. ; Bertram, Torsten
Author_Institution :
Inst. of Control Theor. & Syst. Eng., Tech. Univ. Dortmund, Dortmund, Germany
Abstract :
This paper proposes a novel approach for a derandomized covariance matrix adaptation for multi-objective optimization. Common derandomized multi-objective algorithms only utilize the information gained from successful mutations. However in case of optimization problems with a limited budget for fitness evaluations inferior mutations provide additional information to adjust the search. The proposed algorithm, called active-(μ+λ)-MO-CMA-ES, extends previous approaches as it reduces the covariance along directions of unsuccessful mutations. In experiments on a set of commonly accepted multi-objective test problems the presented algorithm outperforms other derandomized evolution strategies.
Keywords :
covariance matrices; evolutionary computation; optimisation; active covariance matrix adaptation; active-(μ+λ)-MO-CMA-ES; derandomized covariance matrix adaptation; derandomized evolution strategy; derandomized multiobjective algorithms; fitness evaluations inferior mutations; multiobjective CMA-ES; multiobjective optimization; Evolutionary computation;
Conference_Titel :
Advanced Computational Intelligence (ICACI), 2013 Sixth International Conference on
Conference_Location :
Hangzhou
Print_ISBN :
978-1-4673-6341-9
DOI :
10.1109/ICACI.2013.6748499