Algorithm Structure Design for Evolutionary Multi-Objective Local Search

2015 Fiscal Year Final Research Report

• Project/Area Number: 24300090
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Partial Multi-year Fund
• Section: 一般
• Research Field: Sensitivity informatics/Soft computing
• Research Institution: Osaka Prefecture University
• Principal Investigator: Ishibuchi Hisao 大阪府立大学, 工学(系)研究科(研究院), 教授 (60193356)
• Project Period (FY): 2012-04-01 – 2016-03-31
• Keywords: アルゴリズム / 多目的最適化 / 遺伝的アルゴリズム / 進化計算 / 遺伝的局所探索

Outline of Final Research Achievements

When we search for a final single solution of a multi-objective optimization problem with conflicting objectives, we need to identify the tradeoff relation among the conflicting objectives. For this purpose, it is important to find a set of so-called Pareto optimal solutions. This is because the set of all Pareto optimal solutions, which is called the Pareto front, shows the tradeoff relation among the conflicting objectives in the objective space. An evolutionary multi-objective optimization (EMO) algorithm tries to find all Pareto optimal solutions by its single run using its multi-point search property. In this project, we have examined how to combine local search with an EMO algorithm for drastically improving its search ability. After examining various schemes for combining local search, we obtained some important guidelines for implementing high-performance evolutionary multi-objective local search.

Free Research Field

計算知能