TL;DR: VariBAD opens a path to tractable approximate Bayes-optimal exploration for deep RL using ideas from meta-learning, Bayesian RL, and approximate variational inference.
Abstract: Trading off exploration and exploitation in an unknown environment is key to maximising expected return during learning. A Bayes-optimal policy, which does so optimally, conditions its actions not only on the environment state but on the agent’s uncertainty about the environment. Computing a Bayes-optimal policy is however intractable for all but the smallest tasks. In this paper, we introduce variational Bayes-Adaptive Deep RL (variBAD), a way to meta-learn to perform approximate inference in an unknown environment, and incorporate task uncer- tainty directly during action selection. In a grid-world domain, we illustrate how variBAD performs structured online exploration as a function of task uncertainty. We further evaluate variBAD on MuJoCo domains widely used in meta-RL and show that it achieves higher online return than existing methods.
Keywords: Meta-Learning, Bayesian Reinforcement Learning, BAMDPs, Deep Reinforcement Learning
Code: [![github](/images/github_icon.svg) lmzintgraf/varibad](https://github.com/lmzintgraf/varibad) + [![Papers with Code](/images/pwc_icon.svg) 2 community implementations](https://paperswithcode.com/paper/?openreview=Hkl9JlBYvr)
Community Implementations: [![CatalyzeX](/images/catalyzex_icon.svg) 2 code implementations](https://www.catalyzex.com/paper/arxiv:1910.08348/code)
Original Pdf: pdf