A1300
Title: Regularization in the face of uncertainty
Authors: Yuejie Chi - Yale University (United States) [presenting]
Abstract: A key challenge in reinforcement learning (RL) is how to tame uncertainty in a practically-implementable and theoretically-grounded form, one that is amenable in the presence of complex function approximation such as large foundation models. We develop both model-based and model-free frameworks that incentivize exploration via regularization, and show they provably achieve the same rates as their standard RL counterparts, bypassing the need of sophisticated uncertainty quantification.
