Course Title: Robot Learning: From Fundamentals to Foundation Models
Semester: Spring 2026
Lecturer : Oier Mees
Teaching assistants : Alexey Gavryushin, Jonas Pai, Liam Achenbach, Nicola Irmiger, Tianxu An, Šimon Sukup, Nicole Damblon, Zador Pataki, Carl Brandner, Aristotelis Sympetheros, Rohan Walia, Rajiv Bharadwaj, David Hohenstatt, Huanyu Guo
Catalogue Link: 263-5911-00L
Lecture: Mon 16:15-18:00, room NO C 60.
- Format: Each session begins with a lecture on the core topic, followed by a paper discussion led by students. On selected weeks, we will also host short guest lectures from experts in the field to conclude the session.
Thursday Practice: Thu 10:15-12:00, room CHN D 29, D 42, D 46, D 48 and IFW A 32.1.
Course GitHub: mees-robot-learning-course/ethz-course-2026
Course Objectives :
This course provides a comprehensive introduction to modern robot learning, combining classical techniques with the latest advances in large-scale models: Students will start by learning the fundamentals of imitation learning, reinforcement learning, and policy optimization, and gradually progress to advanced topics including Vision-Language-Action (VLA) models and foundation models for robotics The objectives of this course are:
- Understand the core principles of imitation learning, reinforcement learning, and policy learning.
- Implement basic robot learning systems in simulation and on real robots.
- Explore state-of-the-art Vision-Language Action and foundation models for robotics.
- Design and evaluate scalable robot learning pipelines integrating perception, control, and multi-modal reasoning.
Examination:
- Paper Presentation & Moderation (Group): 20 %
- Practical Homework (Coding Assignments): 40 %
- Final Project (Group): 40 %
Lecture Tentative Schedule
| Week | Monday | Paper Discussion | Guest Spotlight |
|---|---|---|---|
| Week 1: Feb 16 | Introduction to Robot Learning Slides Recording | No paper discussion. | - |
| Week 2: Feb 23 | Robot Control & MDPs Slides Recording | Simple random search provides a competitive approach to RL (Mania et al., 2018), Deep RL Doesn’t Work Yet (Irpan, 2018), Curiosity-driven Exploration by Self-supervised Prediction (Pathak et al., 2017) | Abishek Gupta (Prof. University of Washington) (confirmed) |
| Week 3: Mar 02 | Imitation Learning Slides Recording | Causal Confusion in IL (Den Haan et al., 2019), The surprising effectiveness of representation learning for visual imitation (Pari et al. 2021), Transporter Networks: Rearranging the Visual World for Robotic Manipulation (Zeng et al., 2020) | Danfei Xu (Prof. Georgia Tech) (confirmed) |
| Week 4: Mar 09 | Reinforcement Learning I | Evolution Strategies as a Scalable Alternative to RL (Salimans et al., 2017), Learning Synergies between Pushing and Grasping (Zeng et al., 2018), Precise and Dexterous Robotic Manipulation via Human-in-the-Loop Reinforcement Learning (Luo et al., 2024) | Aviral Kumar (Prof. Carnegie Mellon University & Google DeepMind) (confirmed) |
| Week 5: Mar 16 | Reinforcement Learning II | End-to-End Training of Deep Visuomotor Policies (Levine et al., 2015), Eureka: Human-Level Reward Design via Coding LLMs (Ma et al., 2023), Latent Plans for Task Agnostic Offline Reinforcement Learning (Rosete-Beas et al., 2022) | Andrew Wagenmaker (Postdoc UC Berkeley) (confirmed) |
| Week 6: Mar 23 | Generative Models | Planning with Diffusion for Flexible Behavior Synthesis (Janner & Du et al., 2022), Implicit Behavioral Cloning (Florence et al., 2021), Steering Your Diffusion Policy with Latent Space RL (Wagenmaker et al., 2025) | Cheng Chi (Co-Founder Sunday Robotics, Lead of Diffusion Policy & UMI) (confirmed) |
| Week 7: Mar 30 | Sequence Modeling and Transformers | Decision Transformer: RL via Sequence Modeling (Chen et al., 2021), Learning Fine-Grained Bimanual Manipulation (ALOHA) (Zhao et al., 2023), Humanoid Locomotion as Next Token Prediction (Radosavovic et al., 2024) | Ted Xiao (Co-Founder Prometheus, ex-Google) (confirmed) |
| Week 8: Apr 13 | World Models | Learning Universal Policies via Text-Guided Video Generation (Du et al, 2023), Training Agents Inside of Scalable World Models (Hafner et al., 2025), World Action Models are Zero-shot Policies (Ye et al., 2026) | Scott Reed (Principal Research Scientist NVIDIA GEAR Lab) (confirmed) |
| Week 9: Apr 27 | Generalist Robot Policies | Language Conditioned Imitation Learning over Unstructured Data (Lynch et al., 2021), A Generalist Agent (Gato) (Reed et al., 2022), π∗0.6: a VLA That Learns From Experience (Physical Intelligence, 2025) | Quan Vuong (Co-Founder Physical Intelligence) (confirmed) |
| Week 10: May 04 | Embodied Reasoning and Test-time Scaling | In-Context Imitation Learning via Next-Token Prediction (Fu et al., 2024), VOYAGER: An Open-Ended Embodied Agent with LLMs (Wang et al., 2023), Training Strategies for Efficient Embodied Reasoning (Chen et al., 2025) | Archit Sharma (confirmed) (Research Scientist Google DeepMind, Co-creator Gemini Deep Think series) |
| Week 11: May 11 | Frontier & Open Problem | A Path Towards Autonomous Machine Intelligence (LeCun, 2022), The Bitter Lesson (Sutton, 2019), Intelligence without Representation (Brooks, 1991) | Lucas Beyer (Meta) (tentative) |
| Week 12: May 18 | Guest Lectures | Dieter Fox (Prof. University of Washington & Director AI2) (confirmed) & Pieter Abbeel (Prof. UC Berkeley & VP Amazon) (tentative) | - |
Tutorials
Each session is organized as follows. TAs first give a summary of the relevant course content and introduce the exercises. TAs then remain in the room to assist students in solving the exercises.
| Week | Topic | Material | Due Date |
|---|---|---|---|
| Week 1: Feb 19 | Pytorch & Numpy Tutorial | Code | March 5 |
| Week 2: Feb 26 | Robot Control & MDPs | Code | March 12 |