GENERATION OF PARETO FRONTIERS USING SUPPORT VECTOR MACHINE
Yeboon Yun, Hirotaka Nakayama, Masaaki Arakawa · 2004
Summary: Approximation methods using computational intelligence, for example, evolutionary algorithms have been applied to multi-objective optimization problems. Those methods have been improved increasingly in order to generate more exactly a lot of approximate Pareto optimal solutions. This paper proposes a new method using support vector machine to find an approximate Pareto frontier in multi-objective optimization problems. Furthermore, this paper shows that combining the proposed method and evolutionary algorithm can generate well approximate Pareto frontier, and a decision making with two or three objective functions can be easily performed on the basis of visualized Pareto frontiers by the proposed method. Finally, the effectiveness of the proposed method will be illustrated through several numerical examples.