Machine Learning for Complex-Structured Data

2019 Fiscal Year Final Research Report

• Project/Area Number: 15H01704
• Research Category: Grant-in-Aid for Scientific Research (A)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Intelligent informatics
• Research Institution: Kyoto University
• Principal Investigator: Kashima Hisashi 京都大学, 情報学研究科, 教授 (80545583)
• Project Period (FY): 2015-04-01 – 2020-03-31
• Keywords: 人工知能 / 機械学習 / データサイエンス / ヒューマンコンピュテーション / クラウドソーシング

Outline of Final Research Achievements

We studied effective data analysis methods for complex-structured data such as graph-structured data and human-generated data. In particular, we developed new models and techniques for deep learning for graphs. On the other hand, we developed various methods to integrate human judges to make better decisions on difficult problems that machine intelligence alone cannot solve. We also applied the developed methods to various scientific applications in chemistry, biology, and material science, and showed their promise.

Free Research Field

人工知能

Academic Significance and Societal Importance of the Research Achievements

本研究は、近年学術界・産業界において期待されている深層学習などの機械学習・データ解析技術の適用可能性を大きく広げるとともに、人工知能だけでは解決が困難な問題に対して人間の知能や判断を組み合わせることで、より適切な意思決定を行うことを可能とするための基盤技術を開発するものである。