mlsanityData to 14 Oct 2026

PEnGUiN: Partially Equivariant Graph NeUral Networks for Sample Efficient MARL

arXiv 2503.15615 ↗First posted 19 Mar 2025cs.LG cs.AI cs.ROcited by 2

Not on the map. Why?

Abstract (arXiv)

Equivariant Graph Neural Networks (EGNNs) have emerged as a promising approach in Multi-Agent Reinforcement Learning (MARL), leveraging symmetry guarantees to greatly improve sample efficiency and generalization. However, real-world environments often exhibit inherent asymmetries arising from factors such as external forces, measurement inaccuracies, or intrinsic system biases. This paper introduces textitPartially Equivariant Graph NeUral Networks (PEnGUiN), a novel architecture specifically designed to address these challenges. We formally identify and categorize various types of partial equivariance relevant to MARL, including subgroup equivariance, feature-wise equivariance, regional equivariance, and approximate equivariance. We theoretically demonstrate that PEnGUiN is capable of learning both fully equivariant (EGNN) and non-equivariant (GNN) representations within a unified framework. Through extensive experiments on a range of MARL problems incorporating various asymmetries, we empirically validate the efficacy of PEnGUiN. Our results consistently demonstrate that PEnGUiN outperforms both EGNNs and standard GNNs in asymmetric environments, highlighting their potential to improve the robustness and applicability of graph-based MARL algorithms in real-world scenarios.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 22 of its 26 references that are in the map. First posted on arXiv 19 Mar 2025.

Builds on

22 of 26 references.

References · 22

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2017Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environmentscited by 1,558
  3. 2021E(n) Equivariant Graph Neural Networkscited by 965
  4. 2021Geometric and Physical Quantities Improve E(3) Equivariant Message Passingcited by 228
  5. 2021A Practical Method for Constructing Equivariant Multilayer Perceptrons for Arbitrary Matrix Groupscited by 187
  6. 2020MDP Homomorphic Networks: Group Symmetries in Reinforcement Learningcited by 142
  7. 2022Approximately Equivariant Networks for Imperfectly Symmetric Dynamicscited by 97
  8. 2021Residual Pathway Priors for Soft Equivariance Constraintscited by 75
  9. 2023Approximation-Generalization Trade-offs under (Approximate) Group Equivariancecited by 46
  10. 2022Relaxing Equivariance Constraints with Non-stationary Continuous Filterscited by 41
  11. 2022The Surprising Effectiveness of Equivariant Models in Domains with Latent Symmetrycited by 33
  12. 2021Multi-Agent MDP Homomorphic Networkscited by 25
10 more
  1. 2023Discovering Symmetry Breaking in Physical Systems with Relaxed Group Convolutioncited by 20
  2. 2023A General Theory of Correct, Incorrect, and Extrinsic Equivariancecited by 19
  3. 2023Approximately Equivariant Graph Networkscited by 16
  4. 2023Subequivariant Graph Reinforcement Learning in 3D Environmentscited by 11
  5. 2023Leveraging Partial Symmetry for Multi-Agent Reinforcement Learningcited by 11
  6. 2023Almost Equivariance via Lie Algebra Convolutionscited by 10
  7. 2024Boosting Sample Efficiency and Generalization in Multi-agent Reinforcement Learning via Equivariancecited by 10
  8. 2024Approximate Equivariance in Reinforcement Learningcited by 9
  9. 2024Relaxed Equivariant Graph Neural Networkscited by 7
  10. 2024Symmetry-Based Structured Matrices for Efficient Approximately Equivariant Networkscited by 7

Built on by

Cited by year

Citing papers per year: 2025: 2.

2 papers.

  1. 2025cMALC-D: Contextual Multi-Agent LLM-Guided Curriculum Learning with Diversity-Based Context Blendingcited by 2
  2. 2025Local-Canonicalization Equivariant Graph Neural Networks for Sample-Efficient and Generalizable Swarm Robot Controlcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Equivariant Reinforcement Learning under Partial Observability
  2. 2023Equivariant MuZero
  3. 2025Partially Equivariant Reinforcement Learning in Symmetry-Breaking Environments
  4. 2024Equivariant Offline Reinforcement Learning
  5. 2025SE(3)-Equivariant Robot Learning and Control: A Tutorial Survey
  6. 2024Equivariant Ensembles and Regularization for Reinforcement Learning in Map-based Path Planning
  7. 2025Reinforcement Networks: novel framework for collaborative Multi-Agent Reinforcement Learning tasks
  8. 2025Combining Local Symmetry Exploitation and Reinforcement Learning for Optimised Probabilistic Inference -- A Work In Progress