Antti Oulasvirta (Aalto University): Better AI with Better Human Models

Time and place:
December 2nd, 14:15
Lecture room T5, CS Buidling, Aalto University (in person) and Zoom

Abstract: The ability to estimate the state of a human partner is an insufficient basis on which to build cooperative agents. Also needed is an ability to predict how people adapt their behavior in response to an agent's actions. We propose a new approach based on computational rationality, which models humans based on the idea that predictions can be derived by calculating policies that are approximately optimal given human-like bounds. Computational rationality brings together reinforcement learning and cognitive modeling in pursuit of this goal, facilitating machine understanding of humans.

Speaker:  Antti Oulasvirta leads the Computational Behavior Lab (cbl.aalto.fi) at Aalto University and the Interactive AI research program at FCAI (Finnish Center for AI). Prior to joining Aalto, he was a Senior Researcher at the Max Planck Institute for Informatics. He received his doctorate in Cognitive Science from the University of Helsinki in 2006, after which he was a Fulbright Scholar at the School of Information in University of California-Berkeley in 2007-2008 and a Senior Researcher at Helsinki Institute for Information Technology HIIT in 2008-2011. He was awarded the ERC Starting Grant (2015-2020) for research on computational design of user interfaces and the ERC Advanced Grant (2024-2029) for studying computational models of human behavior. Dr. Oulasvirta serves as an associate editor for ACM TOCHI and has previously served International Journal of Human-Computer Studies, as well as served as a column editor for IEEE Computer. He His work has been awarded the Best Paper Award and Best Paper Honorable Mention at CHI fifteen times between 2008 and 2024. He is a Fellow of ELLIS (European Laboratory for Learning and Intelligent Systems).