mlsanityData to 14 Oct 2026

Multi-Group Equivariant Augmentation for Reinforcement Learning in Robot Manipulation

arXiv 2508.11204 ↗First posted 15 Aug 2025cs.RO cs.AIcited by 0

On the map: Deep reinforcement learning methods → RL for robot locomotion and manipulation

Abstract (arXiv)

Sampling efficiency is critical for deploying visuomotor learning in real-world robotic manipulation. While task symmetry has emerged as a promising inductive bias to improve efficiency, most prior work is limited to isometric symmetries -- applying the same group transformation to all task objects across all timesteps. In this work, we explore non-isometric symmetries, applying multiple independent group transformations across spatial and temporal dimensions to relax these constraints. We introduce a novel formulation of the partially observable Markov decision process (POMDP) that incorporates the non-isometric symmetry structures, and propose a simple yet effective data augmentation method, Multi-Group Equivariance Augmentation (MEA). We integrate MEA with offline reinforcement learning to enhance sampling efficiency, and introduce a voxel-based visual representation that preserves translational equivariance. Extensive simulation and real-robot experiments across two manipulation domains demonstrate the effectiveness of our approach.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods. Builds on 44 of its 49 references that are in the map. First posted on arXiv 15 Aug 2025.

Builds on

44 of 49 references.

Closest ancestors · 3

Same direction, newest first

  1. 2024Equivariant Reinforcement Learning under Partial Observability26 Aug
  2. 2024Revisiting Data Augmentation in Deep Reinforcement Learning19 Feb
  3. 2024SERL: A Software Suite for Sample-Efficient Robotic Reinforcement Learning29 Jan

Other references · 41

  1. 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,887
  2. 2023Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardwarecited by 1,568
  3. 2017Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real Worldcited by 1,498
  4. 2016Group Equivariant Convolutional Networkscited by 1,165
  5. 2020Mastering Atari with Discrete World Modelscited by 1,007
  6. 2015End-to-End Training of Deep Visuomotor Policiescited by 875
  7. 2020CURL: Contrastive Unsupervised Representations for Reinforcement Learningcited by 760
  8. 2018QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulationcited by 743
  9. 2022R3M: A Universal Visual Representation for Robot Manipulationcited by 694
  10. 2020Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixelscited by 585
  11. 2020Reinforcement Learning with Augmented Datacited by 468
  12. 2022DayDreamer: World Models for Physical Robot Learningcited by 400
29 more
  1. 2018Quantifying Generalization in Reinforcement Learningcited by 352
  2. 2020Transporter Networks: Rearranging the Visual World for Robotic Manipulationcited by 344
  3. 2017Asymmetric Actor Critic for Image-Based Robot Learningcited by 238
  4. 2016CAD2RL: Real Single-Image Flight without a Single Real Imagecited by 233
  5. 2021Neural Descriptor Fields: SE(3)-Equivariant Object Representations for Manipulationcited by 209
  6. 2020Generalization in Reinforcement Learning by Soft Data Augmentationcited by 137
  7. 2021Recurrent Model-Free RL Can Be a Strong Baseline for Many POMDPscited by 122
  8. 2019Network Randomization: A Simple Technique for Generalization in Deep Reinforcement Learningcited by 104
  9. 2022SO(2)-Equivariant Reinforcement Learningcited by 84
  10. 2016Multi-view Self-supervised Deep Learning for 6D Pose Estimation in the Amazon Picking Challengecited by 70
  11. 2021Equivariant Q Learning in Spatial Action Spacescited by 62
  12. 2022Equivariant Descriptor Fields: SE(3)-Equivariant Energy-Based Models for End-to-End Visual Robotic Manipulation Learningcited by 56
  13. 2022Sample Efficient Grasp Learning Using Equivariant Modelscited by 53
  14. 2023Diffusion-EDFs: Bi-equivariant Denoising Generative Modeling on SE(3) for Visual Robotic Manipulationcited by 52
  15. 2022SE(3)-Equivariant Relational Rearrangement with Neural Descriptor Fieldscited by 51
  16. 2022Equivariant Transporter Networkcited by 46
  17. 2024DextrAH-RGB: Visuomotor Policies to Grasp Anything with Dexterous Handscited by 45
  18. 2022On-Robot Learning With Equivariant Modelscited by 42
  19. 2022Edge Grasp Network: A Graph-Based SE(3)-invariant Approach to Grasp Detectioncited by 35
  20. 2024Fourier Transporter: Bi-Equivariant Robotic Manipulation in 3Dcited by 33
  21. 2022MIRA: Mental Imagery for Robotic Affordancescited by 26
  22. 2022A Comprehensive Survey of Data Augmentation in Visual Reinforcement Learningcited by 25
  23. 2023Local Neural Descriptor Fields: Locally Conditioned Object Representations for Manipulationcited by 21
  24. 2023Leveraging Symmetries in Pick and Placecited by 19
  25. 2024Imagination Policy: Using Generative Point Cloud Models for Learning Manipulation Policiescited by 18
  26. 2022Data Augmentation for Manipulationcited by 16
  27. 2023One-shot Imitation Learning via Interaction Warpingcited by 16
  28. 2024RiEMann: Near Real-Time SE(3)-Equivariant Robot Manipulation without Point Cloud Segmentationcited by 12
  29. 2022BulletArm: An Open-Source Robotic Manipulation Benchmark and Learning Frameworkcited by 9

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Sample Complexity of Equivariant Reinforcement Learning
  2. 2025SE(3)-Equivariant Robot Learning and Control: A Tutorial Survey
  3. 2025Partially Equivariant Reinforcement Learning in Symmetry-Breaking Environments
  4. 2026EquiBim: Learning Symmetry-Equivariant Policy for Bimanual Manipulation
  5. 2026MIRROR: Reflecting Hidden Symmetries from Offline Robot Interaction Data ↗
  6. 2022SEIL: Simulation-augmented Equivariant Imitation Learning
  7. 2025Equivariant Goal Conditioned Contrastive Reinforcement Learning
  8. 2025Eq.Bot: Enhance Robotic Manipulation Learning via Group Equivariant Canonicalization