IFML Seminar
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...
Upcoming Events
- October312:15 - 1:15pm
IFML Seminar
Event DetailsAbstract: We introduce a method for learning behavioral foundation models using the successor measure. We show that any visitation distribution...
- October1012:15 - 1:15pm
IFML Seminar
Event DetailsABSTRACT: What would it mean for AI to understand skilled human activity? In augmented reality (AR), a person wearing smart...
- October2412:15 - 1:15pm
IFML Seminar
Event DetailsAbstract: Learning from sequential, temporally-correlated data is a core facet of modern machine learning and statistical modeling. Yet our fundamental...
Past Events
- April1112:15 - 1:15pm
IFML Seminar
Event DetailsAbstract: The widespread deployment of AI systems in critical domains demands more rigorous approaches to evaluating their capabilities and safety...
- April412:15 - 1:15pm
IFML Seminar
Event DetailsAbstract: Self-play has powered breakthroughs in two-player and multiplayer games. In this talk, I show that self-play is a surprisingly...
- March2812:15 - 1:15pm
IFML Seminar
Event DetailsAbstract: In this talk, we first discuss a new one-shot lower bound on the Bayes error in simple binary hypothesis...
- March1412:15 - 1:15pm
IFML Seminar
Event DetailsAbstract: Diffusion-based generative models transform random noise into images; their inversion aims to transform images back to structured noise that...
-
IFML Seminar
Event DetailsAbstract: I will present a polynomial time algorithm to recover planted cliques in the semi random models introduced by Feige...
- March712:15 - 1:15pm
IFML Seminar
Event DetailsAbstract: In the first part of this presentation, I'll present a non-asymptotic bound for the spectral norm of a Gaussian...
