mlsanityData to 14 Oct 2026

SeMAIL: Eliminating Distractors in Visual Imitation via Separated Models

arXiv 2306.10695 ↗First posted 19 Jun 2023cs.LG cs.AI cs.CVcited by 10

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

Abstract (arXiv)

Model-based imitation learning (MBIL) is a popular reinforcement learning method that improves sample efficiency on high-dimension input sources, such as images and videos. Following the convention of MBIL research, existing algorithms are highly deceptive by task-irrelevant information, especially moving distractors in videos. To tackle this problem, we propose a new algorithm - named Separated Model-based Adversarial Imitation Learning (SeMAIL) - decoupling the environment dynamics into two parts by task-relevant dependency, which is determined by agent actions, and training separately. In this way, the agent can imagine its trajectories and imitate the expert behavior efficiently in task-relevant state space. Our method achieves near-expert performance on various visual control tasks with complex observations and the more challenging tasks with different backgrounds from expert observations.

Key facts. Cited by 10 papers 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 38 of its 49 references that are in the map. First posted on arXiv 19 Jun 2023.

Builds on

38 of 49 references.

Closest ancestors · 10

Same direction, newest first

  1. 2023Visual Imitation Learning with Patch Rewards2 Feb
  2. 2022Watch and Match: Supercharging Imitation with Regularized Optimal Transport30 Jun
  3. 2021Visual Adversarial Imitation Learning using Variational Models16 Jul
  4. 2021Domain-Robust Visual Imitation Learning with Mutual Information Constraints8 Mar
  5. 2019A Divergence Minimization Perspective on Imitation Learning Methods6 Nov
  6. 2019Task-Relevant Adversarial Imitation Learning2 Oct
  7. 2019Imitation Learning from Video by Leveraging Proprioception22 May
  8. 2018An Algorithmic Perspective on Imitation Learning16 Nov
  9. 2018Variational Discriminator Bottleneck: Improving Imitation Learning, Inverse RL, and GANs by Constraining Information Flow1 Oct
  10. 2018Discriminator-Actor-Critic: Addressing Sample Inefficiency and Reward Bias in Adversarial Imitation Learning9 Sep

Other references · 28

  1. 2017Mask R-CNNcited by 7,277
  2. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,316
  3. 2017Improved Training of Wasserstein GANscited by 2,235
  4. 2017The Kinetics Human Action Video Datasetcited by 1,909
  5. 2016OpenAI Gymcited by 1,736
  6. 2019Dream to Control: Learning Behaviors by Latent Imaginationcited by 1,653
  7. 2016Generative Adversarial Imitation Learningcited by 1,597
  8. 2018Learning Latent Dynamics for Planning from Pixelscited by 1,264
  9. 2020Mastering Atari with Discrete World Modelscited by 1,007
  10. 2018DeepMind Control Suitecited by 956
  11. 2020Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixelscited by 585
  12. 2020Learning Invariant Representations for Reinforcement Learning without Reconstructioncited by 436
16 more
  1. 2016Guided Cost Learning: Deep Inverse Optimal Control via Policy Optimizationcited by 406
  2. 2017Inverse Reward Designcited by 249
  3. 2022Masked Visual Pre-training for Motor Controlcited by 248
  4. 2017One-Shot Visual Imitation Learning via Meta-Learningcited by 228
  5. 2019Provably efficient RL with Rich Observations via Latent State Decodingcited by 169
  6. 2022Model-Based Imitation Learning for Urban Drivingcited by 150
  7. 2018Zero-Shot Visual Imitationcited by 117
  8. 2020Visual Imitation Made Easycited by 111
  9. 2019Reward Tampering Problems and Solutions in Reinforcement Learning: A Causal Influence Diagram Perspectivecited by 108
  10. 2022Reinforcement Learning with Action-Free Pre-Training from Videoscited by 108
  11. 2017Third-Person Imitation Learningcited by 103
  12. 2022Denoised MDPs: Learning World Models Better Than the World Itselfcited by 68
  13. 2021Learning Task Informed Abstractionscited by 58
  14. 2022Iso-Dream: Isolating and Leveraging Noncontrollable Visual Dynamics in World Modelscited by 53
  15. 2016Model-based Adversarial Imitation Learningcited by 29
  16. 2022INFOrmation Prioritization through EmPOWERment in Visual Model-Based RLcited by 27

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 4; 2025: 3; 2026: 2.

10 papers.

  1. 2025Latent Action Learning Requires Supervision in the Presence of Distractorscited by 44
  2. 2024AD3: Implicit Action is the Key for World Models to Distinguish the Diverse Visual Distractorscited by 10
  3. 2023Robust Visual Imitation Learning with Inverse Dynamics Representationscited by 8
  4. 2025FOUNDER: Grounding Foundation Models in World Models for Open-Ended Embodied Decision Makingcited by 8
  5. 2024Focus On What Matters: Separated Models For Visual-Based RL Generalizationcited by 6
  6. 2025Reimagination with Test-time Observation Interventions: Distractor-Robust World Model Predictions for Visual Model Predictive Controlcited by 6
  7. 2024Stem-OB: Generalizable Visual Imitation Learning with Stem-Like Convergent Observation through Diffusion Inversioncited by 4
  8. 2026Temporally Centered SIGReg Improves LeWorldModel Representations for Robot Policy Learningcited by 4
  9. 2024SENSOR: Imitate Third-Person Expert's Behaviors via Active Sensoringcited by 1
  10. 2026Beyond Conservatism: Recoverability-Conditioned Exploration for Model-Based Imitation Learningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023Adversarial Imitation Learning from Visual Observations using Latent Information
  2. 2018Generative Adversarial Imitation from Observation
  3. 2022Adversarial Imitation Learning from Video using a State Observer
  4. 2018Sample-Efficient Imitation Learning via Generative Adversarial Nets
  5. 2019Provably Efficient Imitation Learning from Observation Alone
  6. 2026Latent Wasserstein Adversarial Imitation Learning
  7. 2021A Pragmatic Look at Deep Imitation Learning
  8. 2022Resolving Copycat Problems in Visual Imitation Learning via Residual Action Prediction