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.
@article{arxiv.2408.14336,
title = {Equivariant Reinforcement Learning under Partial Observability},
author = {Hai Nguyen and Andrea Baisero and David Klee and Dian Wang and Robert Platt and Christopher Amato},
journal= {arXiv preprint arXiv:2408.14336},
year = {2024}
}