A study on EPR method for solving large scale combinatorial optimization problems

2015 Fiscal Year Final Research Report

• Project/Area Number: 25330262
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Intelligent informatics
• Research Institution: Kyushu University
• Principal Investigator: Fujita Hiroshi 九州大学, システム情報科学研究院, 准教授 (70284552)
• Co-Investigator(Kenkyū-buntansha): HASEGAWA Ryuzo 九州大学, 大学院システム情報科学研究院, 教授 (20274483) ; KOSHIMURA Miyuki 九州大学, 大学院システム情報科学研究院, 助教 (30274492)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Keywords: 組合せ最適化 / Ramsey数 / SAT / MaxSAT / 基数制約

Outline of Final Research Achievements

We have developed methods for solving large scale combinatorial problems. As a result, SCSat system which is based on SAT techniques enhanced with a mechanism to utilize soft constraints for effective solution search, succeeded to solve several hard mathematical problems on Ramsey numbers including the new result R(4,11)>=101.
Also we have developed several ways to encode cardinality constraints into SAT problems that are used in solving large MaxSAT problems. Equipped with the encoding mechanism, QMaxSAT system has been improved so that much larger problems can be solved than before, and achieved excellent results in international competitions on systems of this kind. Our successful case studies include security solutions for cryptography and an alternative implementation of inductive logic programming.

Free Research Field

知能情報学