Improving Generalization in Meta Reinforcement Learning using Learned Objectives

Louis Kirsch, Sjoerd van Steenkiste, Juergen Schmidhuber

TL;DR: We introduce MetaGenRL, a novel meta reinforcement learning algorithm. Unlike prior work, MetaGenRL can generalize to new environments that are entirely different from those used for meta-training.

Abstract: Biological evolution has distilled the experiences of many learners into the general learning algorithms of humans. Our novel meta reinforcement learning algorithm MetaGenRL is inspired by this process. MetaGenRL distills the experiences of many complex agents to meta-learn a low-complexity neural objective function that decides how future individuals will learn. Unlike recent meta-RL algorithms, MetaGenRL can generalize to new environments that are entirely different from those used for meta-training. In some cases, it even outperforms human-engineered RL algorithms. MetaGenRL uses off-policy second-order gradients during meta-training that greatly increase its sample efficiency.

Keywords: meta reinforcement learning, meta learning, reinforcement learning