Collaborative Research: CIF: Medium: Learning to Control from Data: from Theory to Practice

The goal of this project is to develop a new generation of data-driven decision-making theory and algorithms to address pressing challenges in modern Reinforcement Learning (RL). (i) Specifically, it aims to design sample-efficient and computationally efficient algorithms for online and offline RL under general function approximation. (ii) Also, it aims to improve the adaptivity and trustworthiness of the existing RL training paradigm. PI Wang from Northwestern will focus on the first goal.