2026-09-19T09:26:31Zhttps://dspace.mit.edu/server/oai/requestoai:dspace.mit.edu:1721.1/1555082024-07-09T03:05:57Zcom_1721.1_7582com_1721.1_7581col_1721.1_131022
Sontag, David
Satyanarayan, Arvind
Glassman, Elena
Horvitz, Eric
Mozannar, Hussein
Massachusetts Institute of Technology. Institute for Data, Systems, and Society
2024-07-08T18:55:54Z
2024-07-08T18:55:54Z
2024-05
2024-06-03T20:46:12.190Z
https://hdl.handle.net/1721.1/155508
AI systems are augmenting humans' capabilities in settings such as healthcare and programming, forming human-AI teams. To enable more accurate and timely decisions, we need to optimize the performance of the human-AI team directly. In this thesis, we utilize a mathematical framing of the human-AI team and propose a set of methods that optimize the AI, the human, and the interface in which they communicate to enable better team performance. We first show how to provably train AI classifiers that complement humans and can defer the decision to humans when it is best to do so. However, in specific settings, AI cannot autonomously make decisions and thus only provides advice to humans. In that case, we build onboarding procedures that train humans to have an accurate mental model of the AI to enable appropriate reliance. Finally, we study how humans interact with large language models (LLMs) to write code. To understand current inefficiencies, we developed a taxonomy to categorize programmers' interactions with the LLM. Motivated by insight from the taxonomy, we leverage human feedback to know when to best display LLM suggestions.
Ph.D.
Massachusetts Institute of Technology
Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)
Copyright retained by author(s)
https://creativecommons.org/licenses/by-nc-nd/4.0/
Training Human-AI Teams
Thesis
application/pdf
Doctoral
Doctor of Philosophy
Publication
unknown
unknown
<Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="dbec5244-b0cd-4309-9053-34482ba03643">
<Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type>
<Title>Training Human-AI Teams</Title>
<PublishedIn>
<Publication>
</Publication>
</PublishedIn>
<PublicationDate>2024-05</PublicationDate>
<Authors>
<Author>
<DisplayName>Mozannar, Hussein</DisplayName>
<Affiliation>
<OrgUnit>
</OrgUnit>
</Affiliation>
</Author>
</Authors>
<Editors>
</Editors>
<Publishers>
<Publisher>
<DisplayName>Massachusetts Institute of Technology</DisplayName>
<OrgUnit />
</Publisher>
</Publishers>
<License>https://creativecommons.org/licenses/by-nc-nd/4.0/</License>
<Abstract>AI systems are augmenting humans' capabilities in settings such as healthcare and programming, forming human-AI teams. To enable more accurate and timely decisions, we need to optimize the performance of the human-AI team directly. In this thesis, we utilize a mathematical framing of the human-AI team and propose a set of methods that optimize the AI, the human, and the interface in which they communicate to enable better team performance. We first show how to provably train AI classifiers that complement humans and can defer the decision to humans when it is best to do so. However, in specific settings, AI cannot autonomously make decisions and thus only provides advice to humans. In that case, we build onboarding procedures that train humans to have an accurate mental model of the AI to enable appropriate reliance. Finally, we study how humans interact with large language models (LLMs) to write code. To understand current inefficiencies, we developed a taxonomy to categorize programmers' interactions with the LLM. Motivated by insight from the taxonomy, we leverage human feedback to know when to best display LLM suggestions.</Abstract>
<Access xmlns="http://purl.org/coar/access_right"
>
</Access>
</Publication>