Equivariant Reinforcement Learning under Partial Observability

2024/08/26 by Hai V. Nguyen, Andrea Baisero, Nguyen, Hai +9 · 6 citations
Computer Science · Engineering · #Artificial Intelligence (cs.AI) #Computer Vision and Pattern Recognition (cs.CV) #Elevator Systems and Control #FOS: Computer and information sciences #Fault Detection and Control Systems #Robotics (cs.RO) #Software Reliability and Analysis Research

paper · pdf · doi:10.48550/arxiv.2408.14336

openalex publication_date 2024/08/26 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

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.

Cited by

• EfficientFlow: Efficient Equivariant Flow Policy Learning for Embodied AI
• Partially Equivariant Reinforcement Learning in Symmetry-Breaking Environments
• Reinforcement Learning Using known Invariances
• Local-Canonicalization Equivariant Graph Neural Networks for Sample-Efficient and Generalizable Swarm Robot Control
• Multi-Group Equivariant Augmentation for Reinforcement Learning in Robot Manipulation
• Constraint-Preserving Data Generation for Visuomotor Policy Learning

Related

• SO(2)-Equivariant Reinforcement Learning
• Group Equivariant Deep Reinforcement Learning
• Multi-Group Equivariant Augmentation for Reinforcement Learning in Robot Manipulation
• The Surprising Effectiveness of Equivariant Models in Domains with Latent Symmetry
• Partially Equivariant Reinforcement Learning in Symmetry-Breaking Environments