mlsanityData to 7 Oct 2026

Learning Multi-Stage Tasks with One Demonstration via Self-Replay

arXiv 2111.07447 ↗First posted 14 Nov 2021cs.RO cs.LGcited by 21

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

Abstract (arXiv)

In this work, we introduce a novel method to learn everyday-like multi-stage tasks from a single human demonstration, without requiring any prior object knowledge. Inspired by the recent Coarse-to-Fine Imitation Learning method, we model imitation learning as a learned object reaching phase followed by an open-loop replay of the demonstrator's actions. We build upon this for multi-stage tasks where, following the human demonstration, the robot can autonomously collect image data for the entire multi-stage task, by reaching the next object in the sequence and then replaying the demonstration, and then repeating in a loop for all stages of the task. We evaluate with real-world experiments on a set of everyday-like multi-stage tasks, which we show that our method can solve from a single demonstration. Videos and supplementary material can be found at https://www.robot-learning.uk/self-replay.

Key facts. Cited by 21 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 27 of its 35 references that are in the map. First posted on arXiv 14 Nov 2021.

Builds on

27 of 35 references.

Closest ancestors · 6

Same direction, newest first

  1. 2021Coarse-to-Fine Imitation Learning: Robot Manipulation from a Single Demonstration13 May
  2. 2019Relay Policy Learning: Solving Long-Horizon Tasks via Imitation and Reinforcement Learning25 Oct
  3. 2019Self-Supervised Correspondence in Visuomotor Policy Learning16 Sep
  4. 2019Continuous Relaxation of Symbolic Planner for One-Shot Imitation Learning16 Aug
  5. 2018One-Shot Hierarchical Imitation Learning of Compound Visuomotor Tasks25 Oct
  6. 2018Task-Embedded Control Networks for Few-Shot Imitation Learning8 Oct

Other references · 21

  1. 2016Generative Adversarial Imitation Learningcited by 1,576
  2. 2015End-to-End Training of Deep Visuomotor Policiescited by 847
  3. 2017Learning Robust Rewards with Adversarial Inverse Reinforcement Learningcited by 485
  4. 2018An Algorithmic Perspective on Imitation Learningcited by 455
  5. 2020Transporter Networks: Rearranging the Visual World for Robotic Manipulationcited by 342
  6. 2017Overcoming Exploration in Reinforcement Learning with Demonstrationscited by 322
  7. 2019Learning Latent Plans from Playcited by 310
  8. 2017Leveraging Demonstrations for Deep Reinforcement Learning on Robotics Problems with Sparse Rewardscited by 302
  9. 2017Deep Imitation Learning for Complex Manipulation Tasks from Virtual Reality Teleoperationcited by 286
  10. 2017One-Shot Imitation Learningcited by 232
  11. 2017One-Shot Visual Imitation Learning via Meta-Learningcited by 225
  12. 2019A Divergence Minimization Perspective on Imitation Learning Methodscited by 188
9 more
  1. 2017DART: Noise Injection for Robust Imitation Learningcited by 187
  2. 2018Learning by Playing - Solving Sparse Reward Tasks from Scratchcited by 164
  3. 2015Deep Spatial Autoencoders for Visuomotor Learningcited by 149
  4. 2018One-Shot Imitation from Observing Humans via Domain-Adaptive Meta-Learningcited by 146
  5. 2017Vision-Based Multi-Task Manipulation for Inexpensive Robots Using End-To-End Learning from Demonstrationcited by 100
  6. 2017Neural Task Programming: Learning to Generalize Across Hierarchical Taskscited by 75
  7. 2020FlowControl: Optical Flow Based Visual Servoingcited by 15
  8. 2020Modeling Long-horizon Tasks as Sequential Interaction Landscapescited by 14
  9. 2020SAFARI: Safe and Active Robot Imitation Learning with Imaginationcited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 1; 2022: 3; 2023: 5; 2024: 7; 2025: 4; 2026: 1.

21 papers.

Sort:

By area:

  1. 2023MimicGen: A Data Generation System for Scalable Robot Learning using Human Demonstrationscited by 308
  2. 2024OKAMI: Teaching Humanoid Robots Manipulation Skills through Single Video Imitationcited by 62
  3. 2023HYDRA: Hybrid Robot Actions for Imitation Learningcited by 54
  4. 2024Robot See Robot Do: Imitating Articulated Object Manipulation with Monocular 4D Reconstructioncited by 51
  5. 2024SkillMimicGen: Automated Demonstration Generation for Efficient Skill Learning and Deploymentcited by 48
  6. 2024Vision-based Manipulation from Single Human Video with Open-World Object Graphscited by 46
  7. 2024DINOBot: Robot Manipulation via Retrieval and Alignment with Vision Foundation Modelscited by 43
  8. 2022Demonstrate Once, Imitate Immediately (DOME): Learning Visual Servoing for One-Shot Imitation Learningcited by 36
  9. 2023One-Shot Imitation Learning: A Pose Estimation Perspectivecited by 21
  10. 2023On the Effectiveness of Retrieval, Alignment, and Replay in Manipulationcited by 15
  11. 2025Multi-Stage Manipulation with Demonstration-Augmented Reward, Policy, and World Model Learningcited by 10
  12. 2025DemoHLM: From One Demonstration to Generalizable Humanoid Loco-Manipulationcited by 10
9 more
  1. 2024MILES: Making Imitation Learning Easy with Self-Supervisioncited by 9
  2. 2023NOD-TAMP: Generalizable Long-Horizon Planning with Neural Object Descriptorscited by 8
  3. 2025Subtask-Aware Visual Reward Learning from Segmented Demonstrationscited by 7
  4. 2021Back to Reality for Imitation Learningcited by 3
  5. 2022Conditional Visual Servoing for Multi-Step Taskscited by 3
  6. 2025HybridGen: VLM-Guided Hybrid Planning for Scalable Data Generation of Imitation Learningcited by 3
  7. 2022All You Need is LUV: Unsupervised Collection of Labeled Images using Invisible UV Fluorescent Indicatorscited by 2
  8. 2024Foundation Models for Rapid Autonomy Validationcited by 0
  9. 2026Easy-IIL: Reducing Human Operational Burden in Interactive Imitation Learning via Assistant Expertscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2020Scalable Multi-Task Imitation Learning with Autonomous Improvement
  2. 2021Towards More Generalizable One-shot Visual Imitation Learning
  3. 2024Robotic Imitation of Human Actions
  4. 2020Learning to Generalize Across Long-Horizon Tasks from Human Demonstrations
  5. 2024VITAL: Interactive Few-Shot Imitation Learning via Visual Human-in-the-Loop Corrections
  6. 2025Learning a Thousand Tasks in a Day
  7. 2026Human-to-Robot Interaction: Learning from Video Demonstration for Robot Imitation
  8. 2019To Follow or not to Follow: Selective Imitation Learning from Observations