자료유형  

 학위논문

Control Number  

0017163433

International Standard Book Number  

9798383698587

Dewey Decimal Classification Number  

620

Main Entry-Personal Name  

Dunning, Richard E.

Publication, Distribution, etc. (Imprint  

[S.l.] : Carnegie Mellon University., 2024

Publication, Distribution, etc. (Imprint  

Ann Arbor : ProQuest Dissertations & Theses, 2024

Physical Description  

305 p.

General Note  

Source: Dissertations Abstracts International, Volume: 86-02, Section: A.

General Note  

Advisor: Fischhoff, Baruch.

Dissertation Note  

Thesis (Ph.D.)--Carnegie Mellon University, 2024.

Summary, Etc.  

요약Many applications of AI require humans and AI advisors to make decisions collaboratively; however, success depends on how appropriately humans rely on the AI agent. We demonstrated an evaluation method for a platform that used neural network agents of varying skill levels for the simple strategic game of Connect Four. We manipulated the presence, sequence, skill, and information display of Artificial Intelligence (AI) advice in a strategy game against another AI opponent that sometimes varied its skill to measure their effect on users' performance.Human agent teams outperformed unaided subjects with those receiving the AI recommendations simultaneously achieving the best results. Although team performance was higher and subjects improved during game play, there was little evidence of learning from their AI advisors. AI reliability proved to be the greatest determiner of team performance with subjects retaining trust in higher skilled advisors even in varied environments. Those with higher numeracy demonstrated the highest ability to make use of AI advice including more detailed output formats including ranking of choices and probabilities. More reliable AI agents correlated to higher AI trust while higher self-confidence correlated to greater rejection of AI advice, greater confidence in success, but slightly lower performance.The value of these human agent teams depended on AI reliability, users' ability to extract lessons from their advice, and users' trust in that advice. Organizations implementing human agent teams should conduct testing to know how well users appropriately rely on AI recommendations.

Subject Added Entry-Topical Term  

Engineering.

Subject Added Entry-Topical Term  

Public policy.

Index Term-Uncontrolled  

Appropriate reliance

Index Term-Uncontrolled  

Decision science

Index Term-Uncontrolled  

Human agent teams

Index Term-Uncontrolled  

Trust

Added Entry-Corporate Name  

Carnegie Mellon University Engineering and Public Policy

Host Item Entry  

Dissertations Abstracts International. 86-02A.

Electronic Location and Access  

로그인을 한후 보실 수 있는 자료입니다.

Control Number  

joongbu:655914