Offline Imitation Learning Reading List

Curated by Mouhssine Rifaki | Stanford Electrical Engineering | Last updated August 2026

Key papers on offline imitation learning — learning to imitate from fixed datasets without environment interaction, via distribution matching, conservative objectives, and reward-free frameworks.

Offline Imitation Learning: 10 key papers

1. Imitation Learning from Observations by Minimizing Inverse Dynamics Disagreement
   Yang et al. arXiv 2019.
2. Discriminator-Weighted Offline Imitation Learning from Suboptimal Demonstrations
   Xu et al. arXiv 2022.
3. Imitating Unknown Policies via Exploration
   Gavenski et al. arXiv 2020.
4. Robust Imitation Learning from Noisy Demonstrations
   Tangkaratt et al. arXiv 2020.
5. Off-Policy Imitation Learning from Observations
   Zhu et al. arXiv 2021.
6. Curriculum Offline Imitation Learning
   Liu et al. arXiv 2021.
7. When Should We Prefer Offline Reinforcement Learning Over Behavioral Cloning?
   Kumar et al. arXiv 2022.
8. Model-Based Imitation Learning for Urban Driving
   Hu et al. arXiv 2022.
9. Imitation Learning by Estimating Expertise of Demonstrators
   Beliaev et al. arXiv 2022.
10. Robust Imitation of a Few Demonstrations with a Backwards Model
    Park and Wong. arXiv 2022.

Related reading lists

• Imitation Learning and Inverse RL
• Behavioral Cloning & Offline Imitation
• Offline Reinforcement Learning
• Inverse Reinforcement Learning
• Off-Policy Evaluation & Offline Evaluation
• Robust & Risk-Sensitive RL
• Adversarial Reinforcement Learning & Robustness to Adversaries
• Offline-to-Online RL & Fine-Tuning