Imitation Learning Resources

A structured, step-by-step learning guide for understanding Imitation Learning (IL), from foundational concepts to the most relevant algorithms and modern frontiers, with a focus on robotics applications. This repository was built for researchers and practitioners in Reinforcement Learning for Robotics who want to develop a deep understanding of imitation learning. Rather than a flat list of links, this is a curated curriculum: resources are sequenced intentionally, so each step builds on the previous one.

1. Building the Reinforcement Learning/MDP Foundation

🎬 Lectures

David Silver's RL Course (DeepMind x UCL) 2018

📖 Book

"Reinforcement Learning: An Introduction" by Richard S. Sutton and Andrew G. Barto

2. Understanding Why Imitation Learning Exists

🎬 Lectures

Berkeley CS285 - Lecture 2: Imitation Learning by Sergey Levine - Part 1

Berkeley CS285 - Lecture 2: Imitation Learning by Sergey Levine - Part 2

Berkeley CS285 - Lecture 2: Imitation Learning by Sergey Levine - Part 3

Berkeley CS285 - Lecture 2: Imitation Learning by Sergey Levine - Part 4

3. Learning the Core Algorithms: Behavoral Cloning (BC)

🎬 Lectures

Berkeley CS285 Lecture 2 (mentioned above) covers BC formally

🔗 Supplementary

University of Waterloo CS885 - Module 3: Imitation Learning by Pascal Poupart

📖 Book

"MIT Underactuated Robotics" by Russ Tedrake - Chapter 21: Imitation Learning"

4. Learning the Core Algorithms: DAgger

📄 Paper

Ross, S., Gordon, G., & Bagnell, D. (2011). A reduction of imitation learning and structured prediction to no-regret online learning

🎬 Lectures

Berkeley CS285 Lecture 2 (mentioned above) covers DAgger formally

🔗 Supplementary

Cornell University CS6756 - Lecture 8: DAgger and Interactive Experts by Sanjiban Choudhury

5. Learning the Core Algorithms: Inverse Reinforcement Learning (IRL)

🎬 Lectures

Berkeley CS285 - Lecture 20: Inverse Reinforcement Learning by Sergey Levine - Part 1

Berkeley CS285 - Lecture 20: Inverse Reinforcement Learning by Sergey Levine - Part 2

Berkeley CS285 - Lecture 20: Inverse Reinforcement Learning by Sergey Levine - Part 3

Berkeley CS285 - Lecture 20: Inverse Reinforcement Learning by Sergey Levine - Part 4

🔗 Supplementary

Foundations of Deep RL by Pieter Abbeel

6. Learning the Core Algorithms: Generative Adversarial Imitation Learning (GAIL)

📄Paper

Ho, J., & Ermon, S. (2016). Generative adversarial imitation learning

🎬 Lectures

Berkeley CS285 Lecture 20 (mentioned above) covers GAIL and the connection between IRL and GANs formally

🔗 Supplementary

University of Waterloo CS885 - Module 3: Imitation Learning by Pascal Poupart

7. The Robotics-Specific Context

🎬 Lectures

MIT 6.832 Underactuated Robotics, Spring 2024 - Lecture 23: Imitation Learning

📖 Book

"Algorithms for Decision Making" by Kochenderfer, Wheeler & Wray - Chapter 18: Imitation Learning

8. Modern Frontiers (WIP)

📄 Papers

1. Torabi, F., Warnell, G., & Stone, P. (2018). Behavioral cloning from observation
2. Chi, C., Xu, Z., Feng, S., Cousineau, E., Du, Y., Burchfiel, B., ... & Song, S. (2025). Diffusion policy: Visuomotor policy learning via action diffusion