Project/Area Number |
17K01246
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Multi-year Fund |
Section | 一般 |
Research Field |
Social systems engineering/Safety system
|
Research Institution | Tokyo University of Agriculture and Technology |
Principal Investigator |
Miyashiro Ryuhei 東京農工大学, 工学(系)研究科(研究院), 准教授 (50376860)
|
Project Period (FY) |
2017-04-01 – 2020-03-31
|
Project Status |
Completed (Fiscal Year 2019)
|
Budget Amount *help |
¥4,680,000 (Direct Cost: ¥3,600,000、Indirect Cost: ¥1,080,000)
Fiscal Year 2019: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2018: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2017: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
|
Keywords | OR / 整数計画法 / 数理計画法 / アルゴリズム / 最適化 / 数理計画 |
Outline of Final Research Achievements |
Mathematical optimization problems are not only theoretically but also practically important. These problems contain a class of nonconvex discrete optimization problems. Such problems are difficult to solve exactly because of their nonlinear and discrete nature. However, recent progress on machine learning requests us to solve these problems exactly. In this research, we developed a model building method and algorithms via integer programming (integer optimization) for a class of nonconvex discrete optimization problems. The proposed method allows us to solve feature selection problems in statistics and machine learning faster than existing methods. In addition, the quality of the obtained solutions is better than those obtained by the previous methods, for example, L1-regularization.
|
Academic Significance and Societal Importance of the Research Achievements |
社会的に解きたい問題には,複数のものからいくつかのものを選択する際に,最もよい選び方を決めたいという形式のものが多い.ただしこの種の問題は,現在の解き方では高性能なコンピュータを用いても計算に多大な時間がかかるものが大部分である.本研究では,そのような問題のうち特定の形式の問題について,より高速に解くための方法を開発した.研究成果を応用することにより,例えば機械学習という分野で扱われる「特徴選択」と呼ばれるような問題がより高速に解けるようになることが期待できる.
|