Next Seminar

Tuesday 21st MAY 2024, 5PM UTC


Speaker: Ruiqi Zhang (UC Berkeley)

Title: Policy Finetuning in Reinforcement Learning via Design of Experiments using Offline Data

Paper: https://arxiv.org/abs/2307.04354

Slides: TBD

The recording will be uploaded here after the event.

Authors: Ruiqi Zhang, Andrea Zanette

Abstract: In some applications of reinforcement learning, a dataset of pre-collected experience is already available but it is also possible to acquire some additional online data to help improve the quality of the policy. However, it may be preferable to gather additional data with a single, non-reactive exploration policy and avoid the engineering costs associated with switching policies. In this paper we propose an algorithm with provable guarantees that can leverage an offline dataset to design a single non-reactive policy for exploration. We theoretically analyze the algorithm and measure the quality of the final policy as a function of the local coverage of the original dataset and the amount of additional data collected. 

Speaker Bio: 

How To Join: