Equivariant Reinforcement Learning under Partial Observability

Hai Huu Nguyen, Andrea Baisero, David Klee, Dian Wang, Robert Platt, Christopher Amato

Keywords: Partial Observability, Equivariant Learning, Symmetry

TL;DR: This paper embeds domain symmetry in actor-critic reinforcement learning agents to solve a specific class of partially observable domains

Abstract: Incorporating inductive biases is a promising approach for tackling challenging robot learning domains with sample-efficient solutions. This paper identifies partially observable domains where symmetries can be a useful inductive bias for efficient learning. Specifically, by encoding the equivariance regarding specific group symmetries into the neural networks, our actor-critic reinforcement learning agents can reuse solutions in the past for related scenarios. Consequently, our equivariant agents outperform non-equivariant approaches significantly in terms of sample efficiency and final performance, demonstrated through experiments on a range of robotic tasks in simulation and real hardware.

Student First Author: yes

Instructions: I have read the instructions for authors (https://corl2023.org/instructions-for-authors/)

Website: https://sites.google.com/view/equi-rl-pomdp

Code: https://github.com/hai-h-nguyen/equi-rl-for-pomdps