Debiasing Meta-Gradient Reinforcement Learning by Learning the Outer Value Function

1 code implementation • 19 Nov 2022 • Clément Bonnet, Laurence Midgley, Alexandre Laterre

This bias comes from using the critic that is trained using the meta-learned discount factor for the advantage estimation in the outer objective which requires a different discount factor.

reinforcement-learning Reinforcement Learning (RL)