ºÚÁÏÍø

Abstract

Artificial intelligence offers substantial promise, including the promise of automating mundane tasks, improving upon human performance, and augmenting human performance and experience.  While much research focuses on the performance of AI systems in isolation, human-centered AI researchers focus on the importance of the performance and experience of humans engaging with and supported by AI tools.

In this talk I draw lessons from 30 years of work on recommender systems–personalization tools that learn from explicit and implicit human feedback to improve consumer experience when faced with large numbers of choices of products, news, media, etc.  Throughout, I look at the competing roles of offline evaluation and experimentation with users in advancing both algorithmic and human performance.

About the Speaker

Joseph A. Konstan is Distinguished McKnight University Professor and Distinguished University Teaching Professor of Computer Science and Engineering at the University of Minnesota where he also serves as Assoc. Dean for Research in the College of Science and Engineering.  His work focuses on human-centered computing with particular focus on recommender systems and computing applications in behavioral health.  He holds an A.B. from Harvard and an M.S. and Ph.D. from the University of California, Berkeley, all in Computer Science.  He is a Fellow of the ACM, IEEE, and AAAS, a member of the CHI Academy, and winner of the ACM Software System Award.  Prof. Konstan is also known for two open online course sequences on the Coursera platform — one on Recommender Systems and the other on User Interface Design and Evaluation.