Simultaneous Problem Set Optimization Using Evolutionary Computation

2022 Fiscal Year Final Research Report

• Project/Area Number: 19K12135
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 61040:Soft computing-related
• Research Institution: The University of Electro-Communications
• Principal Investigator: Sato Hiroyuki 電気通信大学, 大学院情報理工学研究科, 准教授 (60550978)
• Project Period (FY): 2019-04-01 – 2023-03-31
• Keywords: 最適化 / 進化計算

Outline of Final Research Achievements

We studied evolutionary computation methodology to optimize multiple objective functions simultaneously. For continuous problems with real value design variables and discrete problems with discrete design variables, binary 0/1 especially, we respectively proposed test optimization problems that could specify similarities among objective functions. For these problems, we proposed evolutionary optimization algorithms that estimate similarities among objective functions by solution distributions in the variable space and utilize them to enhance or suppress the cooperative search in crossover generating new solutions by recombining two existing solutions. Results showed that the proposed algorithms could estimate similarities among multiple objective functions and utilized them to improve the simultaneous optimization of multiple objective functions compared to conventional evolutionary algorithms.

Free Research Field

ソフトコンピューティング

Academic Significance and Societal Importance of the Research Achievements

学術的には，異なる複数の目的関数の関係性が未知の状態から，最適化の過程で類似性を推定し，類似度が高い目的関数について，探索する解集合を共有する方法論として意義があると考えられる．社会的には，昨今，設計最適化などに進化計算が利用されるようになってきており，本研究によって，人々の多様なニーズに合わせたバリエーションが豊富な製品を同時に設計するシーンなどへの利用が促進される意義があると考えられる．