Title :
Strength Pareto evolutionary algorithm based on Bayes with decision graphs
Author :
Yao, Jintao ; Lin, Yaping ; Kong, Yuyan ; Zhong, Minjuan
Author_Institution :
Coll. of Comput. Sci. & Commun., Hunan Univ., Changsha, China
Abstract :
This paper presents a multi-objective evolutionary algorithm based on Bayes with decision graphs, namely BSPEA, which replaces crossover and mutation operators in traditional EAs by building and learning Bayesian Networks, and avoids setting a lot of parameters manually and destroying some important building blocks. The simulation results demonstrate that the proposed algorithm is capable of converging on the better Pareto front quickly, and has a strong robustness.
Keywords :
Bayes methods; Pareto optimisation; belief networks; decision theory; evolutionary computation; graph theory; knapsack problems; learning (artificial intelligence); Bayes method; Pareto evolutionary algorithm; decision graphs; knapsack problems; learning Bayesian networks; multiobjective evolutionary algorithm; Bayesian methods; Computer science; Educational institutions; Evolutionary computation; Genetic algorithms; Genetic mutations; Mathematics; Pareto optimization; Robustness; Sorting;
Conference_Titel :
Intelligent Control and Automation, 2004. WCICA 2004. Fifth World Congress on
Print_ISBN :
0-7803-8273-0
DOI :
10.1109/WCICA.2004.1341979