Title of Presentation: On the Role of LLMs in Human Decision Making

Abstract:

The rapid emergence of large language models (LLMs) and Generative AI (GenAI) has fundamentally redefined the broad availability and usage of AI. This talk assesses the critical role GenAI can play in supporting human decision-making. Using conversational recommender systems as a lens, I will detail how LLMs can both enhance the quality of decision outcomes and facilitate significantly more natural and effective user interaction. The discussion will cover key techniques, including LLMs for preference elicitation, decision critiquing, and user simulation. Finally, I will outline the current challenges and future directions required to establish LLMs as even more robust decision-support companions.

Bio

Dr. Boutilier is a Principal Scientist at Google, where he works on decision making under uncertainty, with a particular focus on sequential decision models, including reinforcement learning, Markov decision processes, and temporal reasoning. Prior to joining Google in 2015, he was a Professor in the Department of Computer Science at the University of Toronto and held a Canada Research Chair in Adaptive Decision Making for Intelligent Systems. He received his Ph.D. in Computer Science from the University of Toronto and previously served as a faculty member at the University of British Columbia. He also served as Chair of the Department of Computer Science at the University of Toronto and co-founded Granata Decision Systems.

Over the course of his distinguished career, Dr. Boutilier has held visiting and adjunct

appointments at several leading institutions, including Stanford University, Brown University, Carnegie Mellon University, and Université Paris-Dauphine. He has made significant contributions to the artificial intelligence research community through his editorial and professional service, including serving as Editor-in-Chief of the Journal of Artificial Intelligence Research. He has also organized and led numerous international conferences and workshops in AI.

Dr. Boutilier is a Fellow of the Royal Society of Canada, the Association for Computing

Machinery (ACM) and the Association for the Advancement of Artificial Intelligence (AAAI). He has won several awards, including but not limited to the ACM/SIGAI Autonomous Agents Research Award, a Tier I Canada Research Chair, the Isaac Walton Killam Research Fellowship, the IBM Faculty Award, the Killam Teaching Award, and numerous best paper awards at leading AI conferences and journals.

https://research.google/people/craigboutilier/?&