mlsanityData to 14 Oct 2026

Causal Action Influence Aware Counterfactual Data Augmentation

arXiv 2405.18917 ↗First posted 29 May 2024cs.LG cs.AI cs.ROcited by 12

Not on the map. Why?

Abstract (arXiv)

Offline data are both valuable and practical resources for teaching robots complex behaviors. Ideally, learning agents should not be constrained by the scarcity of available demonstrations, but rather generalize beyond the training distribution. However, the complexity of real-world scenarios typically requires huge amounts of data to prevent neural network policies from picking up on spurious correlations and learning non-causal relationships. We propose CAIAC, a data augmentation method that can create feasible synthetic transitions from a fixed dataset without having access to online environment interactions. By utilizing principled methods for quantifying causal influence, we are able to perform counterfactual reasoning by swapping itaction-unaffected parts of the state-space between independent trajectories in the dataset. We empirically show that this leads to a substantial increase in robustness of offline learning algorithms against distributional shift.

Key facts. Cited by 12 papers in the map (data through 14 Oct 2026). Builds on 48 of its 75 references that are in the map. First posted on arXiv 29 May 2024.

Builds on

48 of 75 references.

References · 48

  1. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 3,080
  2. 2018Addressing Function Approximation Error in Actor-Critic Methodscited by 2,417
  3. 2020FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidencecited by 2,085
  4. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,061
  5. 2017Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real Worldcited by 1,498
  6. 2018Relational inductive biases, deep learning, and graph networkscited by 1,360
  7. 2021Offline Reinforcement Learning with Implicit Q-Learningcited by 1,293
  8. 2018Off-Policy Deep Reinforcement Learning without Explorationcited by 1,241
  9. 2017Hindsight Experience Replaycited by 1,136
  10. 2020Mastering Atari with Discrete World Modelscited by 1,007
  11. 2021A Minimalist Approach to Offline Reinforcement Learningcited by 899
  12. 2020Object-Centric Learning with Slot Attentioncited by 876
36 more
  1. 2019Stabilizing Off-Policy Q-Learning via Bootstrapping Error Reductioncited by 840
  2. 2015Generating Sentences from a Continuous Spacecited by 612
  3. 2020Reinforcement Learning with Augmented Datacited by 468
  4. 2020Learning Invariant Representations for Reinforcement Learning without Reconstructioncited by 436
  5. 2019Relay Policy Learning: Solving Long-Horizon Tasks via Imitation and Reinforcement Learningcited by 384
  6. 2019Multi-Object Representation Learning with Iterative Variational Inferencecited by 360
  7. 2018Quantifying Generalization in Reinforcement Learningcited by 352
  8. 2019Learning Latent Plans from Playcited by 316
  9. 2019Causal Confusion in Imitation Learningcited by 288
  10. 2018Multi-Goal Reinforcement Learning: Challenging Robotics Environments and Request for Researchcited by 260
  11. 2017Asymmetric Actor Critic for Image-Based Robot Learningcited by 238
  12. 2022Bridging the Gap to Real-World Object-Centric Learningcited by 220
  13. 2022Human-to-Robot Imitation in the Wildcited by 161
  14. 2021Discovering and Achieving Goals via World Modelscited by 131
  15. 2019SCALOR: Generative World Models with Scalable Object Representationscited by 108
  16. 2019Network Randomization: A Simple Technique for Generalization in Deep Reinforcement Learningcited by 104
  17. 2020Counterfactual Data Augmentation using Locally Factored Dynamicscited by 93
  18. 2018Woulda, Coulda, Shoulda: Counterfactually-Guided Policy Searchcited by 82
  19. 2022Latent Plans for Task-Agnostic Offline Reinforcement Learningcited by 81
  20. 2023Object-Centric Learning for Real-World Videos by Predicting Temporal Feature Similaritiescited by 78
  21. 2018Confounding-Robust Policy Improvementcited by 76
  22. 2020Fighting Copycat Agents in Behavioral Cloning from Observation Historiescited by 71
  23. 2021Causal Influence Detection for Improving Efficiency in Reinforcement Learningcited by 71
  24. 2021Risk-Averse Offline Reinforcement Learningcited by 65
  25. 2022Causal Dynamics Learning for Task-Independent State Abstractioncited by 65
  26. 2020Sample-Efficient Reinforcement Learning via Counterfactual-Based Data Augmentationcited by 59
  27. 2022Learning Agile Skills via Adversarial Imitation of Rough Partial Demonstrationscited by 45
  28. 2018Deconfounding Reinforcement Learning in Observational Settingscited by 36
  29. 2022Generalizing Goal-Conditioned Reinforcement Learning with Variational Causal Reasoningcited by 34
  30. 2018Contingency-Aware Exploration in Reinforcement Learningcited by 26
  31. 2023Seeing is not Believing: Robust Reinforcement Learning against Spurious Correlationcited by 25
  32. 2020Causally Correct Partial Models for Reinforcement Learningcited by 24
  33. 2021Neuro-algorithmic Policies enable Fast Combinatorial Generalizationcited by 11
  34. 2019Mega-Reward: Achieving Human-Level Play without Extrinsic Rewardscited by 5
  35. 2023Efficient Learning of High Level Plans from Playcited by 2
  36. 2023Can Active Sampling Reduce Causal Confusion in Offline Reinforcement Learning?cited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 2; 2025: 7; 2026: 3.

12 papers.

  1. 2024Dream to Manipulate: Compositional World Models Empowering Robot Imitation Learning with Imaginationcited by 45
  2. 2024Unifying Causal Representation Learning with the Invariance Principlecited by 31
  3. 2025Learning Interactive World Model for Object-Centric Reinforcement Learningcited by 11
  4. 2025Null Counterfactual Factor Interactions for Goal-Conditioned Reinforcement Learningcited by 6
  5. 2025Towards Empowerment Gain through Causal Structure Learning in Model-Based RLcited by 5
  6. 2025Curious Causality-Seeking Agents Learn Meta Causal Worldcited by 4
  7. 2025An Analysis of Causal Effect Estimation using Outcome Invariant Data Augmentationcited by 2
  8. 2025Causal Process Models: Reframing Dynamic Causal Graph Discovery as a Reinforcement Learning Problemcited by 1
  9. 2025CAIMAN: Causal Action Influence Detection for Sample-efficient Loco-manipulationcited by 0
  10. 2026Context-Aware Counterfactual Data Augmentation for Gender Bias Mitigation in Language Modelscited by 0
  11. 2026Decoupling Semantics and Fingerprints: A Universal Representation for AI-Generated Image Detectioncited by 0
  12. 2026PAPO-VLA: Planning-Aware Policy Optimization for Vision-Language-Action Modelscited by 0