IFML Seminar: 10/03/25 - Successor Measures and Self-supervised Reinforcement Learning

Abstract: We introduce a method for learning behavioral foundation models using the successor measure. We show that any visitation distribution can be represented using an affine combination of policy-independent basis functions. By learning these basis functions during a self-supervised pre-training phase, we can zero-shot extract a policy for any downstream task. We then show that many self-supervised RL methods can be unified through the successor measure, providing insights on future research directions. 

Bio: Amy is an assistant professor at UT Austin in the Chandra Family Department of Electrical and Computer Engineering. Her work focuses on improving generalization in reinforcement learning through bridging theory and practice in learning and utilizing structure in real world problems. Previously she was a research scientist at Meta AI - FAIR and a postdoctoral fellow at UC Berkeley. She obtained her PhD from McGill University and the Mila Institute, and also previously obtained an M.Eng. in EECS and dual B.Sci. degrees in Mathematics and EECS from MIT.

Zoom link: https://utexas.zoom.us/j/84254847215