Ensemble Critics, Recurrent Policies, and Image-Based Reinforcement Learning

This project focuses on implementing and analyzing advanced reinforcement learning techniques across three distinct tasks: Specifically, it aims to:
(1) Ensemble Critics for an accurate estimate of the value function,
(2) Recurrent Policies in complex environments with limited observability, and
(3) Image-Based Policy Learning for large-scale inputs.