Human Computation Foundations Based on Deep Modeling of Human Decision Making

2022 Fiscal Year Final Research Report

• Project/Area Number: 18H03337
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Single-year Grants
• Section: 一般
• Review Section: Basic Section 62020:Web informatics and service informatics-related
• Research Institution: Hokkaido University
• Principal Investigator: Oyama Satoshi 北海道大学, 情報科学研究院, 准教授 (30346100)
• Project Period (FY): 2018-04-01 – 2022-03-31
• Keywords: ヒューマンコンピュテーション / 深層学習

Outline of Final Research Achievements

Human computation is an approach that combines the abilities of both computers and humans to solve problems that are difficult to achieve sufficient performance with computers alone and require human recognition and judgment, by considering humans as part of the computational resources. In this study, by introducing state-of-the-art machine learning techniques such as deep learning on the computer side, we have improved the performance of human computation. Furthermore, in human computation, where humans are involved, the black box nature of machine learning is a limitation, so we worked to improve its explanatory and interpretability.

Free Research Field

人工知能

Academic Significance and Societal Importance of the Research Achievements

最近の大規模言語モデルに基づく人工知能の急速な発展に伴い、人間と人工知能を代替的に考えるのではなく、それぞれの長所を活かして協働して問題解決に取り組むというヒューマンコンピュテーションのアプローチの重要性が再認識されている。本研究で取り組んだヒューマンコンピュテーションの性能向上や説明性・解釈性の向上は、今後一層重要性を増すと予想され、その学術的意義や社会的意義は大きいと考えている。