mlsanityData to 14 Oct 2026

IntervenGen: Interventional Data Generation for Robust and Data-Efficient Robot Imitation Learning

arXiv 2405.01472 ↗First posted 2 May 2024cs.RO cs.AIcited by 25

On the map: Offline RL and imitation learning from data → Robot manipulation imitation learning

Abstract (arXiv)

Imitation learning is a promising paradigm for training robot control policies, but these policies can suffer from distribution shift, where the conditions at evaluation time differ from those in the training data. A popular approach for increasing policy robustness to distribution shift is interactive imitation learning (i.e., DAgger and variants), where a human operator provides corrective interventions during policy rollouts. However, collecting a sufficient amount of interventions to cover the distribution of policy mistakes can be burdensome for human operators. We propose IntervenGen (I-Gen), a novel data generation system that can autonomously produce a large set of corrective interventions with rich coverage of the state space from a small number of human interventions. We apply I-Gen to 4 simulated environments and 1 physical environment with object pose estimation error and show that it can increase policy robustness by up to 39x with only 10 human interventions. Videos and more results are available at https://sites.google.com/view/intervengen2024.

Key facts. Cited by 25 papers in the map (data through 14 Oct 2026). Placed in Robot manipulation imitation learning, one of 13 directions in Offline RL and imitation learning from data. Builds on 38 of its 47 references that are in the map. First posted on arXiv 2 May 2024.

Builds on

38 of 47 references.

Closest ancestors · 10

Same direction, newest first

  1. 2023MimicGen: A Data Generation System for Scalable Robot Learning using Human Demonstrations26 Oct
  2. 2023Imitating Task and Motion Planning with Visuomotor Transformers25 May
  3. 2023Teach a Robot to FISH: Versatile Imitation from One Minute of Demonstrations2 Mar
  4. 2022BC-Z: Zero-Shot Task Generalization with Robotic Imitation Learning4 Feb
  5. 2021Error-Aware Imitation Learning from Teleoperation Data for Mobile Manipulation9 Dec
  6. 2021Coarse-to-Fine Imitation Learning: Robot Manipulation from a Single Demonstration13 May
  7. 2021LazyDAgger: Reducing Context Switching in Interactive Imitation Learning31 Mar
  8. 2020Human-in-the-Loop Imitation Learning using Remote Teleoperation12 Dec
  9. 2020Learning to Generalize Across Long-Horizon Tasks from Human Demonstrations13 Mar
  10. 2018RoboTurk: A Crowdsourcing Platform for Robotic Skill Learning through Imitation7 Nov

Other references · 28

  1. 2022Do As I Can, Not As I Say: Grounding Language in Robotic Affordancescited by 2,348
  2. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,061
  3. 2017Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real Worldcited by 1,498
  4. 2021What Matters in Learning from Offline Human Demonstrations for Robot Manipulationcited by 725
  5. 2017Sim-to-Real Transfer of Robotic Control with Dynamics Randomizationcited by 672
  6. 2020robosuite: A Modular Simulation Framework and Benchmark for Robot Learningcited by 511
  7. 2022VIMA: General Robot Manipulation with Multimodal Promptscited by 407
  8. 2017Deep Imitation Learning for Complex Manipulation Tasks from Virtual Reality Teleoperationcited by 289
  9. 2019RoboNet: Large-Scale Multi-Robot Learningcited by 270
  10. 2018HG-DAgger: Interactive Imitation Learning with Human Expertscited by 266
  11. 2021Bridge Data: Boosting Generalization of Robotic Skills with Cross-Domain Datasetscited by 250
  12. 2021MT-Opt: Continuous Multi-Task Robotic Reinforcement Learning at Scalecited by 229
16 more
  1. 2022Interactive Language: Talking to Robots in Real Timecited by 218
  2. 2017DART: Noise Injection for Robust Imitation Learningcited by 191
  3. 2020Recovery RL: Safe Reinforcement Learning with Learned Recovery Zonescited by 175
  4. 2021ThriftyDAgger: Budget-Aware Novelty and Risk Gating for Interactive Imitation Learningcited by 108
  5. 2022Robot Learning on the Job: Human-in-the-Loop Autonomy and Learning During Deploymentcited by 108
  6. 2022You Only Demonstrate Once: Category-Level Manipulation from Single Visual Demonstrationcited by 81
  7. 2021Robust Multi-Modal Policies for Industrial Assembly via Reinforcement Learning and Demonstrations: A Large-Scale Studycited by 30
  8. 2022Fleet-DAgger: Interactive Robot Fleet Learning with Scalable Human Supervisioncited by 30
  9. 2014Interactive Policy Learning through Confidence-Based Autonomycited by 29
  10. 2021Guided Imitation of Task and Motion Planningcited by 18
  11. 2022Leveraging Fully Observable Policies for Learning under Partial Observabilitycited by 18
  12. 2016Comparing Human-Centric and Robot-Centric Sampling for Robot Deep Learning from Demonstrationscited by 17
  13. 2023Diagnosis, Feedback, Adaptation: A Human-in-the-Loop Framework for Test-Time Policy Adaptationcited by 16
  14. 2022Visual Backtracking Teleoperation: A Data Collection Protocol for Offline Image-Based Reinforcement Learningcited by 11
  15. 2016Learning to Gather Information via Imitationcited by 4
  16. 2023Get Back Here: Robust Imitation by Return-to-Distribution Planningcited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 4; 2025: 13; 2026: 8.

25 papers.

Sort:

By area:

  1. 2024DexMimicGen: Automated Data Generation for Bimanual Dexterous Manipulation via Imitation Learningcited by 116
  2. 2024TRANSIC: Sim-to-Real Policy Transfer by Learning from Online Correctioncited by 59
  3. 2024SkillMimicGen: Automated Demonstration Generation for Efficient Skill Learning and Deploymentcited by 52
  4. 2024RACER: Rich Language-Guided Failure Recovery Policies for Imitation Learningcited by 49
  5. 2025RaC: Robot Learning for Long-Horizon Tasks by Scaling Recovery and Correctioncited by 39
  6. 2025Dexterous Manipulation through Imitation Learning: A Surveycited by 21
  7. 2025AirExo-2: Scaling up Generalizable Robotic Imitation Learning with Low-Cost Exoskeletonscited by 18
  8. 2025MILE: Model-based Intervention Learningcited by 10
  9. 2025Constraint-Preserving Data Generation for Visuomotor Policy Learningcited by 6
  10. 2025R2RGEN: Real-to-Real 3D Data Generation for Spatially Generalized Manipulationcited by 6
  11. 2026Data Pyramid for Embodied Manipulation: A Surveycited by 6
  12. 2025FieldGen: From Teleoperated Pre-Manipulation Trajectories to Field-Guided Data Generationcited by 4
13 more
  1. 2025Real2Edit2Real: Generating Robotic Demonstrations via a 3D Control Interfacecited by 4
  2. 2026OHP-RL: Online Human Preference as Guidance in Reinforcement Learning for Robot Manipulationcited by 3
  3. 2025AOMGen: Photoreal, Physics-Consistent Demonstration Generation for Articulated Object Manipulationcited by 2
  4. 2025Safety-Aware Imitation Learning via MPC-Guided Disturbance Injectioncited by 1
  5. 2025Blindfolded Experts Generalize Better: Insights from Robotic Manipulation and Videogamescited by 1
  6. 2025Balance Equation-based Distributionally Robust Offline Imitation Learningcited by 1
  7. 2025Denoising-based Contractive Imitation Learningcited by 0
  8. 2026R2RDreamer: 3D-aware Data Augmentation for Spatially-generalized 2D Manipulation Policiescited by 0
  9. 2026MinInter: Minimizing Trajectory Interpolation During Data Augmentation for Imitation Learningcited by 0
  10. 2026Improving Imitation Learning Efficiency for Manipulation through Geometric Prior Pretrainingcited by 0
  11. 2026ProTracer: Proprioception-Guided Failure Diagnosis in Robot Manipulationcited by 0
  12. 2026TANDEM: Task and Motion Planning with As-Needed Demonstrations for Efficient Vision-Language-Action Model Fine-tuningcited by 0
  13. 2026Kintsugi-VLA: Turning Failed Robot Rollouts into Recovery Data through Interventional Recoverabilitycited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025ReinforceGen: Hybrid Skill Policies with Automated Data Generation and Reinforcement Learning
  2. 2026Scaling Single Human Demonstrations for Imitation Learning using Generative Foundational Models
  3. 2024VITAL: Interactive Few-Shot Imitation Learning via Visual Human-in-the-Loop Corrections
  4. 2026ExpertGen: Scalable Sim-to-Real Expert Policy Learning from Imperfect Behavior Priors
  5. 2019Generative predecessor models for sample-efficient imitation learning
  6. 2025DynaMimicGen: A Data Generation Framework for Robot Learning of Dynamic Tasks
  7. 2026Active Generative Simulation: Meta-Learning What to Simulate for Sim-and-Real Policy Co-Training ↗
  8. 2022Robust Learning from Observation with Model Misspecification