mlsanityData to 7 Oct 2026

EDGI: Equivariant Diffusion for Planning with Embodied Agents

arXiv 2303.12410 ↗First posted 22 Mar 2023cs.LG cs.RO stat.MLcited by 35

On the map: Diffusion models → Diffusion and flow policies for robotics

Abstract (arXiv)

Embodied agents operate in a structured world, often solving tasks with spatial, temporal, and permutation symmetries. Most algorithms for planning and model-based reinforcement learning (MBRL) do not take this rich geometric structure into account, leading to sample inefficiency and poor generalization. We introduce the Equivariant Diffuser for Generating Interactions (EDGI), an algorithm for MBRL and planning that is equivariant with respect to the product of the spatial symmetry group SE(3), the discrete-time translation group Z, and the object permutation group Sn. EDGI follows the Diffuser framework (Janner et al., 2022) in treating both learning a world model and planning in it as a conditional generative modeling problem, training a diffusion model on an offline trajectory dataset. We introduce a new SE(3)xZxSn-equivariant diffusion model that supports multiple representations. We integrate this model in a planning loop, where conditioning and classifier guidance let us softly break the symmetry for specific tasks as needed. On object manipulation and navigation tasks, EDGI is substantially more sample efficient and generalizes better across the symmetry group than non-equivariant models.

Key facts. Cited by 35 papers in the map (data through 7 Oct 2026). Placed in Diffusion and flow policies for robotics, one of 14 directions in Diffusion models. Builds on 41 of its 51 references that are in the map. First posted on arXiv 22 Mar 2023.

Builds on

41 of 51 references.

Closest ancestors · 1

Same direction, newest first

  1. 2022Is Conditional Generative Modeling all you need for Decision-Making?28 Nov

Other references · 40

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,185
  3. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  4. 2017Neural Message Passing for Quantum Chemistrycited by 3,460
  5. 2018Addressing Function Approximation Error in Actor-Critic Methodscited by 2,302
  6. 2015Variational Inference with Normalizing Flowscited by 2,161
  7. 2020Conservative Q-Learning for Offline Reinforcement Learningcited by 1,777
  8. 2017Deep Setscited by 1,394
  9. 2019Normalizing Flows for Probabilistic Modeling and Inferencecited by 1,315
  10. 2018Off-Policy Deep Reinforcement Learning without Explorationcited by 1,227
  11. 2016Group Equivariant Convolutional Networkscited by 1,154
  12. 2021Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gaugescited by 1,131
28 more
  1. 2022Planning with Diffusion for Flexible Behavior Synthesiscited by 1,023
  2. 2021E(n) Equivariant Graph Neural Networkscited by 951
  3. 2022Equivariant Diffusion for Molecule Generation in 3Dcited by 666
  4. 2022GeoDiff: a Geometric Diffusion Model for Molecular Conformation Generationcited by 511
  5. 2022MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate Force Fieldscited by 467
  6. 2018Spherical CNNscited by 452
  7. 2019General E(2)-Equivariant Steerable CNNscited by 441
  8. 2021Equivariant message passing for the prediction of tensorial properties and molecular spectracited by 438
  9. 2016Steerable CNNscited by 298
  10. 2020Equivariant Flows: Exact Likelihood Generative Learning for Symmetric Densitiescited by 248
  11. 2022Riemannian Score-Based Generative Modellingcited by 216
  12. 2021Scalars are universal: Equivariant machine learning, structured like classical physicscited by 127
  13. 2022Equivariant 3D-Conditional Diffusion Models for Molecular Linker Designcited by 121
  14. 2022Riemannian Diffusion Modelscited by 120
  15. 2020Equivariant flow-based sampling for lattice gauge theorycited by 102
  16. 2020Sampling using SU(N) gauge equivariant flowscited by 83
  17. 2022SO(2)-Equivariant Reinforcement Learningcited by 79
  18. 2020Plannable Approximations to MDP Homomorphisms: Equivariance under Actionscited by 61
  19. 2022On-Robot Learning With Equivariant Modelscited by 38
  20. 2020Trajectory Prediction using Equivariant Continuous Convolutioncited by 26
  21. 2020Theoretical Aspects of Group Equivariant Neural Networkscited by 25
  22. 2022Hyperbolic Deep Reinforcement Learningcited by 25
  23. 2022Continuous MDP Homomorphisms and Homomorphic Policy Gradientcited by 24
  24. 2021Equivariant Manifold Flowscited by 20
  25. 2022Equivariant Networks for Zero-Shot Coordinationcited by 16
  26. 2018The Effect of Planning Shape on Dyna-style Planning in High-dimensional State Spacescited by 15
  27. 2021Equivariant Finite Normalizing Flowscited by 11
  28. 2023Equivariant MuZerocited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 7; 2024: 16; 2025: 9; 2026: 3.

35 papers.

Sort:

By area:

  1. 2023SE(3)-Stochastic Flow Matching for Protein Backbone Generationcited by 147
  2. 2023Policy Representation via Diffusion Probability Model for Reinforcement Learningcited by 105
  3. 2023Compositional Abilities Emerge Multiplicatively: Exploring Diffusion Models on a Synthetic Taskcited by 91
  4. 2024EquiBot: SIM(3)-Equivariant Diffusion Policy for Generalizable and Data Efficient Learningcited by 72
  5. 2024Simple Hierarchical Planning with Diffusioncited by 71
  6. 2024Equivariant Diffusion Policycited by 61
  7. 2023Enhancing Deep Reinforcement Learning: A Tutorial on Generative Diffusion Models in Network Optimizationcited by 56
  8. 2023Diffusion-EDFs: Bi-equivariant Denoising Generative Modeling on SE(3) for Visual Robotic Manipulationcited by 51
  9. 2024Does equivariance matter at scale?cited by 49
  10. 2023Geometric Algebra Transformercited by 46
  11. 2025Diffusion Models for Robotic Manipulation: A Surveycited by 39
  12. 2024Scaling Diffusion Policy in Transformer to 1 Billion Parameters for Robotic Manipulationcited by 38
23 more
  1. 2025Generative Trajectory Stitching through Diffusion Compositioncited by 38
  2. 2023Boosting Continuous Control with Consistency Policycited by 36
  3. 2024A Survey of Embodied Learning for Object-Centric Robotic Manipulationcited by 34
  4. 2024Deep Generative Models for Offline Policy Learning: Tutorial, Survey, and Perspectives on Future Directionscited by 20
  5. 2024Diffusion Model for Planning: A Systematic Literature Reviewcited by 17
  6. 2024Discrete Policy: Learning Disentangled Action Space for Multi-Task Robotic Manipulationcited by 17
  7. 2024Morphological Symmetries in Roboticscited by 9
  8. 2024GraphAgent: Agentic Graph Language Assistantcited by 8
  9. 2025GAGrasp: Geometric Algebra Diffusion for Dexterous Graspingcited by 7
  10. 2025SE(3)-Equivariant Diffusion Policy in Spherical Fourier Spacecited by 7
  11. 2024SE(3) Equivariant Ray Embeddings for Implicit Multi-View Depth Estimationcited by 6
  12. 2024Differentiable and Learnable Wireless Simulation with Geometric Transformerscited by 5
  13. 2024On Diffusion Models for Multi-Agent Partial Observability: Shared Attractors, Error Bounds, and Composite Flowcited by 5
  14. 2024The Duality of Generative AI and Reinforcement Learning in Robotics: A Reviewcited by 5
  15. 2025A Practical Guide for Incorporating Symmetry in Diffusion Policycited by 4
  16. 20253D Equivariant Visuomotor Policy Learning via Spherical Projectioncited by 4
  17. 2025Morphologically Symmetric Reinforcement Learning for Ambidextrous Bimanual Manipulationcited by 3
  18. 2025Diffusion Models for Reinforcement Learning: Foundations, Taxonomy, and Developmentcited by 2
  19. 2024Task-Agnostic Pre-training and Task-Guided Fine-tuning for Versatile Diffusion Plannercited by 1
  20. 2025EquiReg: Equivariance Regularized Diffusion for Inverse Problemscited by 1
  21. 2026DEGMC: Denoising Diffusion Models Based on Riemannian Equivariant Group Morphological Convolutionscited by 0
  22. 2026Morphologically Equivariant Flow Matching for Bimanual Mobile Manipulationcited by 0
  23. 2026Terminal Symmetry as a Carrier of Asymmetric Process Knowledge: Statewise Refinement for Anytime Verified Constructioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Symmetry-Aware Steering of Equivariant Diffusion Policies: Benefits and Limits
  2. 2024ET-SEED: Efficient Trajectory-Level SE(3) Equivariant Diffusion Policy
  3. 2025Multi-Group Equivariant Augmentation for Reinforcement Learning in Robot Manipulation
  4. 2025SE(3)-Equivariant Robot Learning and Control: A Tutorial Survey
  5. 2025Toward Embodiment Equivariant Vision-Language-Action Policy
  6. 2025ReSeFlow: Rectifying SE(3)-Equivariant Policy Learning Flows
  7. 2023DiMSam: Diffusion Models as Samplers for Task and Motion Planning under Partial Observability
  8. 2025What Do You Need for Compositional Generalization in Diffusion Planning?