mlsanityData to 7 Oct 2026

Teach a Robot to FISH: Versatile Imitation from One Minute of Demonstrations

arXiv 2303.01497 ↗First posted 2 Mar 2023cs.RO cs.AI cs.CVcited by 74

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

Abstract (arXiv)

While imitation learning provides us with an efficient toolkit to train robots, learning skills that are robust to environment variations remains a significant challenge. Current approaches address this challenge by relying either on large amounts of demonstrations that span environment variations or on handcrafted reward functions that require state estimates. Both directions are not scalable to fast imitation. In this work, we present Fast Imitation of Skills from Humans (FISH), a new imitation learning approach that can learn robust visual skills with less than a minute of human demonstrations. Given a weak base-policy trained by offline imitation of demonstrations, FISH computes rewards that correspond to the "match" between the robot's behavior and the demonstrations. These rewards are then used to adaptively update a residual policy that adds on to the base-policy. Across all tasks, FISH requires at most twenty minutes of interactive learning to imitate demonstrations on object configurations that were not seen in the demonstrations. Importantly, FISH is constructed to be versatile, which allows it to be used across robot morphologies (e.g. xArm, Allegro, Stretch) and camera configurations (e.g. third-person, eye-in-hand). Our experimental evaluations on 9 different tasks show that FISH achieves an average success rate of 93%, which is around 3.8x higher than prior state-of-the-art methods.

Key facts. Cited by 74 papers in the map (data through 7 Oct 2026). Placed in Robot manipulation imitation learning, one of 13 directions in Offline RL and imitation learning from data. Builds on 54 of its 80 references that are in the map. First posted on arXiv 2 Mar 2023.

Builds on

54 of 80 references.

Closest ancestors · 9

Same direction, newest first

  1. 2022Graph Inverse Reinforcement Learning from Diverse Videos28 Jul
  2. 2022Behavior Transformers: Cloning k modes with one stone22 Jun
  3. 2022Dexterous Imitation Made Easy: A Learning-Based Framework for Efficient Dexterous Manipulation24 Mar
  4. 2021The Surprising Effectiveness of Representation Learning for Visual Imitation2 Dec
  5. 2021Towards More Generalizable One-shot Visual Imitation Learning26 Oct
  6. 2021Playful Interactions for Representation Learning19 Jul
  7. 2021Coarse-to-Fine Imitation Learning: Robot Manipulation from a Single Demonstration13 May
  8. 2020Visual Imitation Made Easy11 Aug
  9. 2020State-Only Imitation Learning for Dexterous Manipulation7 Apr

Other references · 45

  1. 2022Generative Adversarial Networkscited by 5,603
  2. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,712
  3. 2015Continuous control with deep reinforcement learningcited by 3,279
  4. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,277
  5. 2016Generative Adversarial Imitation Learningcited by 1,576
  6. 2021What Matters in Learning from Offline Human Demonstrations for Robot Manipulationcited by 699
  7. 2017Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrationscited by 685
  8. 2022R3M: A Universal Visual Representation for Robot Manipulationcited by 678
  9. 2020AWAC: Accelerating Online Reinforcement Learning with Offline Datasetscited by 647
  10. 2017Learning Robust Rewards with Adversarial Inverse Reinforcement Learningcited by 485
  11. 2018Behavioral Cloning from Observationcited by 482
  12. 2021Implicit Behavioral Cloningcited by 442
33 more
  1. 2021Mastering Visual Continuous Control: Improved Data-Augmented Reinforcement Learningcited by 359
  2. 2019Exploring the Limitations of Behavior Cloning for Autonomous Drivingcited by 343
  3. 2017Overcoming Exploration in Reinforcement Learning with Demonstrationscited by 322
  4. 2018Reinforced Cross-Modal Matching and Self-Supervised Imitation Learning for Vision-Language Navigationcited by 304
  5. 2017Soft-DTW: a Differentiable Loss Function for Time-Seriescited by 265
  6. 2022Masked Visual Pre-training for Motor Controlcited by 246
  7. 2018Residual Reinforcement Learning for Robot Controlcited by 242
  8. 2022Real-World Robot Learning with Masked Visual Pre-trainingcited by 241
  9. 2020Learning Agile Robotic Locomotion Skills by Imitating Animalscited by 229
  10. 2018Discriminator-Actor-Critic: Addressing Sample Inefficiency and Reward Bias in Adversarial Imitation Learningcited by 203
  11. 2018Generative Adversarial Imitation from Observationcited by 183
  12. 2018Residual Policy Learningcited by 146
  13. 2020Primal Wasserstein Imitation Learningcited by 120
  14. 2022From Play to Policy: Conditional Behavior Generation from Uncurated Robot Datacited by 103
  15. 2022Jump-Start Reinforcement Learningcited by 102
  16. 2019Recent Advances in Imitation Learning from Observationcited by 92
  17. 2018Multiple Interactions Made Easy (MIME): Large Scale Demonstrations Data for Imitationcited by 76
  18. 2022Watch and Match: Supercharging Imitation with Regularized Optimal Transportcited by 64
  19. 2019Wasserstein Adversarial Imitation Learningcited by 59
  20. 2022Holo-Dex: Teaching Dexterity with Immersive Mixed Realitycited by 52
  21. 2018On the Convergence and Robustness of Training GANs with Regularized Optimal Transportcited by 51
  22. 2021Cross-Domain Imitation Learning via Optimal Transportcited by 49
  23. 2020CO-Optimal Transportcited by 42
  24. 2021Visual Adversarial Imitation Learning using Variational Modelscited by 42
  25. 2020Imitation Learning with Sinkhorn Distancescited by 30
  26. 2020The MAGICAL Benchmark for Robust Imitationcited by 30
  27. 2021VOILA: Visual-Observation-Only Imitation Learning for Autonomous Navigationcited by 27
  28. 2021Residual Reinforcement Learning from Demonstrationscited by 27
  29. 2019Deep Residual Reinforcement Learningcited by 21
  30. 2021Domain-Robust Visual Imitation Learning with Mutual Information Constraintscited by 20
  31. 2020Aligning Time Series on Incomparable Spacescited by 15
  32. 2019Learning Vision-based Flight in Drone Swarms by Imitationcited by 12
  33. 2020Learn by Observation: Imitation Learning for Drone Patrolling from Videos of A Human Navigatorcited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 11; 2024: 27; 2025: 25; 2026: 11.

74 papers.

Sort:

By area:

  1. 2024Diffusion Policy Policy Optimizationcited by 255
  2. 2023RoboCLIP: One Demonstration is Enough to Learn Robot Policiescited by 128
  3. 20243D Diffusion Policy: Generalizable Visuomotor Policy Learning via Simple 3D Representationscited by 115
  4. 2025ASAP: Aligning Simulation and Real-World Physics for Learning Agile Humanoid Whole-Body Skillscited by 104
  5. 2023On Bringing Robots Homecited by 101
  6. 2025DexGraspVLA: A Vision-Language-Action Framework Towards General Dexterous Graspingcited by 87
  7. 2024From Imitation to Refinement -- Residual RL for Precise Assemblycited by 80
  8. 2024Bunny-VisionPro: Real-Time Bimanual Dexterous Teleoperation for Imitation Learningcited by 69
  9. 2024Policy Decorator: Model-Agnostic Online Refinement for Large Policy Modelcited by 60
  10. 2024Vision-based Manipulation from Single Human Video with Open-World Object Graphscited by 46
  11. 2024DINOBot: Robot Manipulation via Retrieval and Alignment with Vision Foundation Modelscited by 43
  12. 2025Residual Off-Policy RL for Finetuning Behavior Cloning Policiescited by 42
38 more
  1. 2023An Unbiased Look at Datasets for Visuo-Motor Pre-Trainingcited by 41
  2. 20253DFlowAction: Learning Cross-Embodiment Manipulation from 3D Flow World Modelcited by 36
  3. 2024AffordDP: Generalizable Diffusion Policy with Transferable Affordancecited by 35
  4. 2025ResMimic: From General Motion Tracking to Humanoid Whole-body Loco-Manipulation via Residual Learningcited by 34
  5. 2023See to Touch: Learning Tactile Dexterity through Visual Incentivescited by 33
  6. 2023MimicTouch: Leveraging Multi-modal Human Tactile Demonstrations for Contact-rich Manipulationcited by 30
  7. 2024Adaptive Mobile Manipulation for Articulated Objects In the Open Worldcited by 30
  8. 2024Bridging the Human to Robot Dexterity Gap through Object-Oriented Rewardscited by 25
  9. 2025Compliant Residual DAgger: Improving Real-World Contact-Rich Manipulation with Human Correctionscited by 25
  10. 2024IntervenGen: Interventional Data Generation for Robust and Data-Efficient Robot Imitation Learningcited by 24
  11. 2024Lessons from Learning to Spin "Pens"cited by 22
  12. 2025RoboCopilot: Human-in-the-loop Interactive Imitation Learning for Robot Manipulationcited by 22
  13. 2023One-Shot Imitation Learning: A Pose Estimation Perspectivecited by 21
  14. 2025Dexterous Manipulation through Imitation Learning: A Surveycited by 18
  15. 2023Reinforcement Learning with Foundation Priors: Let the Embodied Agent Efficiently Learn on Its Owncited by 16
  16. 2024Composing Pre-Trained Object-Centric Representations for Robotics From "What" and "Where" Foundation Modelscited by 16
  17. 2024P3-PO: Prescriptive Point Priors for Visuo-Spatial Generalization of Robot Policiescited by 16
  18. 2025CordViP: Correspondence-based Visuomotor Policy for Dexterous Manipulation in Real-Worldcited by 16
  19. 2025A Smooth Sea Never Made a Skilled SAILOR: Robust Imitation via Learning to Searchcited by 16
  20. 2023What Matters to You? Towards Visual Representation Alignment for Robot Learningcited by 14
  21. 2024Zero-shot Imitation Policy via Search in Demonstration Datasetcited by 13
  22. 2024Learning In-Hand Translation Using Tactile Skin With Shear and Normal Force Sensingcited by 13
  23. 2025Touch begins where vision ends: Generalizable policies for contact-rich manipulationcited by 13
  24. 2024Gaussian Splatting to Real World Flight Navigation Transfer with Liquid Networkscited by 12
  25. 2024Learning Precise, Contact-Rich Manipulation through Uncalibrated Tactile Skinscited by 11
  26. 2023Imitation Learning from Observation with Automatic Discount Schedulingcited by 10
  27. 2026RFS: Reinforcement Learning with Residual Flow Steering for Dexterous Manipulationcited by 10
  28. 2023PEARL: Zero-shot Cross-task Preference Alignment and Robust Reward Learning for Robotic Manipulationcited by 9
  29. 2024MILES: Making Imitation Learning Easy with Self-Supervisioncited by 9
  30. 2024SPIRE: Synergistic Planning, Imitation, and Reinforcement Learning for Long-Horizon Manipulationcited by 8
  31. 2025RUKA: Rethinking the Design of Humanoid Hands with Learningcited by 8
  32. 2025CDP: Towards Robust Autoregressive Visuomotor Policy Learning via Causal Diffusioncited by 8
  33. 2025Spatial-Temporal Aware Visuomotor Diffusion Policy Learningcited by 8
  34. 2024Align Your Intents: Offline Imitation Learning via Optimal Transportcited by 7
  35. 2025Imitation Learning from a Single Temporally Misaligned Videocited by 7
  36. 2025Scaffolding Dexterous Manipulation with Vision-Language Modelscited by 7
  37. 2026FAWAM: Force-Aware World Action Models for Closed-Loop Contact-Rich Manipulationcited by 7
  38. 2024KOI: Accelerating Online Imitation Learning via Hybrid Key-state Guidancecited by 6

50 most cited of 74 All 74 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2024VITAL: Interactive Few-Shot Imitation Learning via Visual Human-in-the-Loop Corrections
  2. 2026Human-to-Robot Interaction: Learning from Video Demonstration for Robot Imitation
  3. 2022Learning and Retrieval from Prior Data for Skill-based Imitation Learning
  4. 2024Towards Diverse Behaviors: A Benchmark for Imitation Learning with Human Demonstrations
  5. 2025Robotic Manipulation via Imitation Learning: Taxonomy, Evolution, Benchmark, and Challenges
  6. 2020Scalable Multi-Task Imitation Learning with Autonomous Improvement
  7. 2021Manipulator-Independent Representations for Visual Imitation
  8. 2023MimicPlay: Long-Horizon Imitation Learning by Watching Human Play