Bayesian Distributional Policy Gradients

Luchen Li, A. Aldo Faisal

Distributional Reinforcement Learning (RL) maintains the entire probability distribution of the reward-to-go, i.e. the return, providing more learning signals that account for the…