mlsanityData to 14 Oct 2026

Learning from Demonstrations via Capability-Aware Goal Sampling

arXiv 2601.08731 ↗First posted 13 Jan 2026cs.AIcited by 1

On the map: Offline RL and imitation learning from data → Imitation learning methods

Abstract (arXiv)

Despite its promise, imitation learning often fails in long-horizon environments where perfect replication of demonstrations is unrealistic and small errors can accumulate catastrophically. We introduce Cago (Capability-Aware Goal Sampling), a novel learning-from-demonstrations method that mitigates the brittle dependence on expert trajectories for direct imitation. Unlike prior methods that rely on demonstrations only for policy initialization or reward shaping, Cago dynamically tracks the agent's competence along expert trajectories and uses this signal to select intermediate steps--goals that are just beyond the agent's current reach--to guide learning. This results in an adaptive curriculum that enables steady progress toward solving the full task. Empirical results demonstrate that Cago significantly improves sample efficiency and final performance across a range of sparse-reward, goal-conditioned tasks, consistently outperforming existing learning from-demonstrations baselines.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Imitation learning methods, one of 13 directions in Offline RL and imitation learning from data. Builds on 47 of its 58 references that are in the map. First posted on arXiv 13 Jan 2026.

Builds on

47 of 58 references.

Closest ancestors · 15

Same direction, newest first

  1. 2024Efficient Imitation Learning with Conservative World Models21 May
  2. 2023DITTO: Offline Imitation Learning with World Models6 Feb
  3. 2021A Pragmatic Look at Deep Imitation Learning4 Aug
  4. 2021Visual Adversarial Imitation Learning using Variational Models16 Jul
  5. 2021What Matters for Adversarial Imitation Learning?1 Jun
  6. 2021Of Moments and Matching: A Game-Theoretic Framework for Closing the Imitation Gap4 Mar
  7. 2020Addressing reward bias in Adversarial Imitation Learning with neutral reward functions20 Sep
  8. 2020Primal Wasserstein Imitation Learning8 Jun
  9. 2019Imitation Learning via Off-Policy Distribution Matching10 Dec
  10. 2019A Divergence Minimization Perspective on Imitation Learning Methods6 Nov
  11. 2019Random Expert Distillation: Imitation Learning via Expert Policy Support Estimation16 May
  12. 2019Go-Explore: a New Approach for Hard-Exploration Problems30 Jan
3 more
  1. 2018Discriminator-Actor-Critic: Addressing Sample Inefficiency and Reward Bias in Adversarial Imitation Learning9 Sep
  2. 2018Sample-Efficient Imitation Learning via Generative Adversarial Nets6 Sep
  3. 2018Behavioral Cloning from Observation4 May

Other references · 32

  1. 2020Conservative Q-Learning for Offline Reinforcement Learningcited by 1,809
  2. 2019Dream to Control: Learning Behaviors by Latent Imaginationcited by 1,653
  3. 2016Generative Adversarial Imitation Learningcited by 1,597
  4. 2023Mastering Diverse Domains through World Modelscited by 1,253
  5. 2019Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learningcited by 1,204
  6. 2020Mastering Atari with Discrete World Modelscited by 1,007
  7. 2017Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrationscited by 715
  8. 2017Learning Robust Rewards with Adversarial Inverse Reinforcement Learningcited by 490
  9. 2023Efficient Online Reinforcement Learning with Offline Datacited by 363
  10. 2017Overcoming Exploration in Reinforcement Learning with Demonstrationscited by 340
  11. 2023ManiSkill2: A Unified Benchmark for Generalizable Manipulation Skillscited by 327
  12. 2018Multi-Goal Reinforcement Learning: Challenging Robotics Environments and Request for Researchcited by 260
20 more
  1. 2023Cal-QL: Calibrated Offline RL Pre-Training for Efficient Online Fine-Tuningcited by 237
  2. 2024ManiSkill3: GPU Parallelized Robotics Simulation and Rendering for Generalizable Embodied AIcited by 229
  3. 2022Masked World Models for Visual Controlcited by 170
  4. 2016A Connection between Generative Adversarial Networks, Inverse Reinforcement Learning, and Energy-Based Modelscited by 168
  5. 2022Goal-Conditioned Reinforcement Learning: Problems and Solutionscited by 158
  6. 2021Discovering and Achieving Goals via World Modelscited by 131
  7. 2018Reinforcement and Imitation Learning for Diverse Visuomotor Skillscited by 126
  8. 2022Jump-Start Reinforcement Learningcited by 112
  9. 2018Learning Montezuma's Revenge from a Single Demonstrationcited by 80
  10. 2022MoDem: Accelerating Visual Model-Based Reinforcement Learning with Demonstrationscited by 62
  11. 2021Replacing Rewards with Examples: Example-Based Policy Search via Recursive Classificationcited by 38
  12. 2023Planning Goals for Explorationcited by 28
  13. 2016Playing Atari Games with Deep Reinforcement Learning and Human Checkpoint Replaycited by 22
  14. 2018Backplay: "Man muss immer umkehren"cited by 21
  15. 2021Curriculum Design for Teaching via Demonstrations: Theory and Applicationscited by 9
  16. 2019Adaptive Curriculum Generation from Demonstrations for Sim-to-Real Visuomotor Controlcited by 8
  17. 2024Learning World Models for Unconstrained Goal Navigationcited by 6
  18. 2021Automatic Curricula via Expert Demonstrationscited by 5
  19. 2022Task Phasing: Automated Curriculum Learning from Demonstrationscited by 2
  20. 2024Exploring the Edges of Latent State Clusters for Goal-Conditioned Reinforcement Learningcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026PHIRL: Aligning Learned Rewards with Task Progress for Inverse Reinforcement Learningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023GO-DICE: Goal-Conditioned Option-Aware Offline Imitation Learning via Stationary Distribution Correction Estimation
  2. 2024Expert Proximity as Surrogate Rewards for Single Demonstration Imitation Learning
  3. 2021Learning from Imperfect Demonstrations from Agents with Varying Dynamics
  4. 2025When a Robot is More Capable than a Human: Learning from Constrained Demonstrators
  5. 2022Robust Imitation of a Few Demonstrations with a Backwards Model
  6. 2025Consistent Zero-Shot Imitation with Contrastive Goal Inference
  7. 2023Cold Diffusion on the Replay Buffer: Learning to Plan from Known Good States
  8. 2022Imitating, Fast and Slow: Robust learning from demonstrations via decision-time planning