Abstract: In this talk I will begin by discussing the need for personalized recommendation in conversational AI assistants and the fundamental challenges that make this problem difficult – both in general and in existing conversational AI architectures. I will next discuss some initial progress my research group has made in addressing these challenges as well as our attempt to understand the broader scope of possible interactions in natural, context-driven conversational recommendation. I will conclude with a summary of potential directions forward in the near-term leveraging recent revolutionary advances in language models such as ChatGPT along with longer-term challenges to overcome in order to fully achieve natural and personalized conversational recommendation interactions.
Bio: Scott Sanner is an Associate Professor in Industrial Engineering and Cross-appointed in Computer Science at the University of Toronto. Scott earned a PhD in Computer Science from the University of Toronto (2008), an MS in Computer Science from Stanford University (2002), and a double BS in Computer Science and Electrical and Computer Engineering from Carnegie Mellon University (1999). Scott’s research focuses on a broad range of AI topics spanning sequential decision-making, (conversational) recommender systems, and applications of machine/deep learning. Scott is currently an Associate Editor for ACM Transactions on Recommender Systems (TORS), the Machine Learning Journal (MLJ), and the Journal of Artificial Intelligence Research (JAIR). Scott was a co-recipient of paper awards from the AI Journal (2014), Transport Research Board (2016), CPAIOR (2018) and a recipient of a Google Faculty Research Award (2020).
Where: Building 145, Room 1.33