AI RESEARCH
Partially Equivariant Reinforcement Learning in Symmetry-Breaking Environments
arXiv CS.LG
•
ArXi:2512.00915v2 Announce Type: replace Group symmetries provide a powerful inductive bias for reinforcement learning (RL), enabling efficient generalization across symmetric states and actions via group-invariant Marko Decision Processes (MDPs). However, real-world environments almost never realize fully group-invariant MDPs; dynamics, actuation limits, and reward design usually break symmetries, often only locally. Under group-invariant Bellman backups for such cases, local symmetry-breaking