Algorithm for Dynamic Multi-Objective Distributed Constraint Optimization

2016 Fiscal Year Final Research Report

• Project/Area Number: 26330268
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Intelligent informatics
• Research Institution: Kobe University
• Principal Investigator: Okimoto Tenda 神戸大学, 海事科学研究科, 准教授 (10632432)
• Co-Investigator(Renkei-kenkyūsha): Inoue Katsumi 国立情報学研究所, 情報学プリンシプル研究系, 教授 (10252321)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: 分散制約最適化 / 多目的分散制約最適化 / パレート最適性

Outline of Final Research Achievements

In this research, the framework of a dynamic multi-objective distributed constraint optimization problem has been investigated. First, several efficient algorithms have been developed for solving a multi-objective distributed constraint optimization problem, namely (i) a complete algorithm which can guarantee to find all Pareto optimal solutions (called Pareto front), (ii) an incomplete algorithm that finds a subset of Pareto front, i.e., the obtained solutions can guarantee Pareto optimality, and (iii) an approximation algorithm. Next, a formal framework for a dynamic multi-objective distributed constraint optimization problem is defined and an efficient algorithm is also proposed. Finally, as an application domain, we apply our results to a team formation problem and a nurse-scheduling problem. In summary, this research can be executed according to the research plan described in my proposal. Also, several articles about this research have been accepted in the top conferences for AI.

Free Research Field

マルチエージェントシステム