Comparison between Lamarckian and Baldwinian repair on multiobjective 01 knapsack problems.pdf

H. Ishibuchi, S. Kaige, and K. Narukawa, “Comparison between Lamarckian and Baldwinian repair on multiobjective 0/1 knapsack problems,” Lecture Notes in Computer Science 3410: Evolutionary Multi- Criterion Optimization, pp. 370-385, Springer, Berlin, March 2005. (Proc. 3rd International Conference on Evolutionary Multi-Criterion Optimization, pp. 370-385, March 9-11, 2005, Guanajuato, Mexico.) Comparison between Lamarckian and Baldwinian Repair on Multiobjective 0/1 Knapsack Problems Hisao Ishibuchi, Shiori Kaige, and Kaname Narukawa Department of Industrial Engineering, Osaka Prefecture University 1-1 Gakuen-cho, Sakai, Osaka, 599-8531, Japan {hisaoi, shiori, kaname}@ie.osakafu-u.ac.jp Abstract. This paper examines two repair schemes (i.e., Lamarckian and Baldwinian) through computational experiments on multiobjective 0/1 knapsack problems. First we compare Lamarckian and Baldwinian with each other. Experimental results show that the Baldwinian repair outperforms the Lamarckian repair. It is also shown that these repair schemes outperform a penalty function approach. Then we examine partial Lamarckianism where the Lamarckian repair is applied to each individual with a prespecified probability. Experimental results show that a so-called 5% rule works well. Finally partial Lamarckianism is compared with an island model with two subpopulations where each island has a different repair scheme. Experimental results show that the island model slightly outperforms the standard single-population model with the 50% partial Lamarckian repair in terms of the diversity of solutions. 1 Introduction Since 1990s, multiobjective 0/1 knapsack problems have been frequently used to evaluate the performance of various multiobjective metaheuristics including evolutionary multiobjective optimization (EMO) algorithms [4-7, 11, 16, 18]. When EMO algorithms are applied to multiobjective 0/1 knapsack problems, unfeasible solutions are often generated by genetic op