mlsanityData to 7 Oct 2026

SiT: Symmetry-Invariant Transformers for Generalisation in Reinforcement Learning

arXiv 2406.15025 ↗First posted 21 Jun 2024cs.LGcited by 3

Not on the map. Why?

Abstract (arXiv)

An open challenge in reinforcement learning (RL) is the effective deployment of a trained policy to new or slightly different situations as well as semantically-similar environments. We introduce Symmetry-Invariant Transformer (SiT), a scalable vision transformer (ViT) that leverages both local and global data patterns in a self-supervised manner to improve generalisation. Central to our approach is Graph Symmetric Attention, which refines the traditional self-attention mechanism to preserve graph symmetries, resulting in invariant and equivariant latent representations. We showcase SiT's superior generalization over ViTs on MiniGrid and Procgen RL benchmarks, and its sample efficiency on Atari 100k and CIFAR10.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 43 of its 70 references that are in the map. First posted on arXiv 21 Jun 2024.

Builds on

43 of 70 references.

References · 43

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  2. 2017Graph Attention Networkscited by 8,854
  3. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,635
  4. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,581
  5. 2022FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awarenesscited by 3,611
  6. 2021Decision Transformer: Reinforcement Learning via Sequence Modelingcited by 1,528
  7. 2018Relational inductive biases, deep learning, and graph networkscited by 1,341
  8. 2019Mastering Atari, Go, Chess and Shogi by Planning with a Learned Modelcited by 1,327
  9. 2016Group Equivariant Convolutional Networkscited by 1,154
  10. 2021E(n) Equivariant Graph Neural Networkscited by 951
  11. 2018DeepMind Control Suitecited by 935
  12. 2018IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architecturescited by 866
31 more
  1. 2021Deep Reinforcement Learning at the Edge of the Statistical Precipicecited by 781
  2. 2017Rainbow: Combining Improvements in Deep Reinforcement Learningcited by 774
  3. 2017Dynamic Routing Between Capsulescited by 689
  4. 2020SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networkscited by 602
  5. 2020Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixelscited by 573
  6. 2019Model-Based Reinforcement Learning for Ataricited by 495
  7. 2019Leveraging Procedural Generation to Benchmark Reinforcement Learningcited by 478
  8. 2020Reinforcement Learning with Augmented Datacited by 452
  9. 2019General E(2)-Equivariant Steerable CNNscited by 441
  10. 2018Invariant and Equivariant Graph Networkscited by 339
  11. 2022Transformers are Sample-Efficient World Modelscited by 317
  12. 2018Towards a Definition of Disentangled Representationscited by 294
  13. 2021Escaping the Big Data Paradigm with Compact Transformerscited by 261
  14. 2021Mastering Atari Games with Limited Datacited by 235
  15. 2021CrossFormer: A Versatile Vision Transformer Hinging on Cross-scale Attentioncited by 139
  16. 2020MDP Homomorphic Networks: Group Symmetries in Reinforcement Learningcited by 133
  17. 2020Generalization in Reinforcement Learning by Soft Data Augmentationcited by 132
  18. 2021Contrastive Behavioral Similarity Embeddings for Generalization in Reinforcement Learningcited by 123
  19. 2021Stabilizing Deep Q-Learning with ConvNets and Vision Transformers under Data Augmentationcited by 101
  20. 2022Approximately Equivariant Networks for Imperfectly Symmetric Dynamicscited by 95
  21. 2022SO(2)-Equivariant Reinforcement Learningcited by 79
  22. 2020On the Benefits of Invariance in Neural Networkscited by 78
  23. 2020Automatic Data Augmentation for Generalization in Deep Reinforcement Learningcited by 71
  24. 2020Group Equivariant Stand-Alone Self-Attention For Visioncited by 63
  25. 2019TorchBeast: A PyTorch Platform for Distributed RLcited by 41
  26. 2019Symmetry-Based Disentangled Representation Learning requires Interaction with Environmentscited by 35
  27. 2022The Surprising Effectiveness of Equivariant Models in Domains with Latent Symmetrycited by 32
  28. 2021Koopman Q-learning: Offline Reinforcement Learning via Symmetries of Dynamicscited by 27
  29. 2022Continuous MDP Homomorphisms and Homomorphic Policy Gradientcited by 24
  30. 2023Look Beneath the Surface: Exploiting Fundamental Symmetry for Sample-Efficient Offline RLcited by 18
  31. 2021Grid-to-Graph: Flexible Spatial Relational Inductive Biases for Reinforcement Learningcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 1; 2025: 0; 2026: 2.

3 papers.

  1. 2024Approximate Equivariance in Reinforcement Learningcited by 9
  2. 2026DreamSAC: Learning Hamiltonian World Models via Symmetry Explorationcited by 1
  3. 2026Sparse Masked Attention Policies for Reliable Generalizationcited by 0