mlsanityData to 7 Oct 2026

On the Guaranteed Almost Equivalence between Imitation Learning from Observation and Demonstration

arXiv 2010.08353 ↗First posted 16 Oct 2020cs.RO cs.LGcited by 6

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

Abstract (arXiv)

Imitation learning from observation (LfO) is more preferable than imitation learning from demonstration (LfD) due to the nonnecessity of expert actions when reconstructing the expert policy from the expert data. However, previous studies imply that the performance of LfO is inferior to LfD by a tremendous gap, which makes it challenging to employ LfO in practice. By contrast, this paper proves that LfO is almost equivalent to LfD in the deterministic robot environment, and more generally even in the robot environment with bounded randomness. In the deterministic robot environment, from the perspective of the control theory, we show that the inverse dynamics disagreement between LfO and LfD approaches zero, meaning that LfO is almost equivalent to LfD. To further relax the deterministic constraint and better adapt to the practical environment, we consider bounded randomness in the robot environment and prove that the optimizing targets for both LfD and LfO remain almost same in the more generalized setting. Extensive experiments for multiple robot tasks are conducted to empirically demonstrate that LfO achieves comparable performance to LfD. In fact, most common robot systems in reality are the robot environment with bounded randomness (i.e., the environment this paper considered). Hence, our findings greatly extend the potential of LfO and suggest that we can safely apply LfO without sacrificing the performance compared to LfD in practice.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Placed in Imitation learning methods, one of 13 directions in Offline RL and imitation learning from data. Builds on 23 of its 61 references that are in the map. First posted on arXiv 16 Oct 2020.

Builds on

23 of 61 references.

Closest ancestors · 6

Same direction, newest first

  1. 2020State-only Imitation with Transition Dynamics Mismatch27 Feb
  2. 2019Imitation Learning from Observations by Minimizing Inverse Dynamics Disagreement10 Oct
  3. 2019Imitation Learning for Human Pose Prediction8 Sep
  4. 2019Adversarial Imitation Learning from Incomplete Demonstrations29 May
  5. 2018Generative Adversarial Imitation from Observation17 Jul
  6. 2018Behavioral Cloning from Observation4 May

Other references · 17

  1. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,635
  2. 2015Continuous control with deep reinforcement learningcited by 3,279
  3. 2015Trust Region Policy Optimizationcited by 2,503
  4. 2017Improved Training of Wasserstein GANscited by 2,223
  5. 2016OpenAI Gymcited by 1,684
  6. 2018Spectral Normalization for Generative Adversarial Networkscited by 1,618
  7. 2016Generative Adversarial Imitation Learningcited by 1,576
  8. 2016End to End Learning for Self-Driving Carscited by 1,000
  9. 2017Deep Reinforcement Learning that Matterscited by 727
  10. 2019Learning agile and dynamic motor skills for legged robotscited by 402
  11. 2020Learning Agile Robotic Locomotion Skills by Imitating Animalscited by 229
  12. 2019Consistency Regularization for Generative Adversarial Networkscited by 134
5 more
  1. 2017Learning human behaviors from motion capture by adversarial imitationcited by 75
  2. 2020An empirical investigation of the challenges of real-world reinforcement learningcited by 75
  3. 2018One-Shot Learning of Multi-Step Tasks from Observation via Activity Localization in Auxiliary Videocited by 16
  4. 2020A Learning-from-Observation Framework: One-Shot Robot Teaching for Grasp-Manipulation-Release Household Operationscited by 7
  5. 2019Ctrl-Z: Recovering from Instability in Reinforcement Learningcited by 6

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 0; 2023: 2; 2024: 1; 2025: 2.

6 papers.

  1. 2023Adversarial Imitation Learning from Visual Observations using Latent Informationcited by 10
  2. 2024Visually Robust Adversarial Imitation Learning from Videos with Contrastive Learningcited by 9
  3. 2021Off-policy Imitation Learning from Visual Inputscited by 2
  4. 2023Quantum Imitation Learningcited by 1
  5. 2025Beyond Domain Randomization: Event-Inspired Perception for Visually Robust Adversarial Imitation from Videoscited by 1
  6. 2025Restoring Noisy Demonstration for Imitation Learning With Diffusion Modelscited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2025Learning from Observation: A Survey of Recent Advances
  2. 2021Off-Policy Imitation Learning from Observations
  3. 2021Learning from Imperfect Demonstrations from Agents with Varying Dynamics
  4. 2024A Dual Approach to Imitation Learning from Observations with Offline Datasets
  5. 2020Causal Transfer for Imitation Learning and Decision Making under Sensor-shift
  6. 2023Imitation Learning from Observation through Optimal Transport
  7. 2023Imitation Learning from Observation with Automatic Discount Scheduling
  8. 2020Concurrent Training Improves the Performance of Behavioral Cloning from Observation