抄録
Offer Organization: Japan Society for the Promotion of Science, System Name: Grants-in-Aid for Scientific Research, Category: Grant-in-Aid for Research Activity Start-up, Fund Type: competitive_research_funding, Overall Grant Amount: - (direct: 2020000, indirect: 606000)
To develop effective multi-objective evolutionary algorithms (MOEAs) for many-objective optimization problems which optimize more than four objective functions simultaneously, in this work we have developed a novel MOEA partially applying Pareto dominance and a method to self-control the dominance area of solutions. Also, to encourage the solutions search in many-objectives problems, we have developed a method of crossover controlling the number of crossed genes. The result of performance verification using benchmark problems revealed that proposed methods significantly improve the search performance of MOEA on many-objective optimization problems.