Welcome to CS 159!
The goal of the class is to bring students up to speed in two topics in modern machine learning research through a series of lectures. Students will then go on to conduct a mini research project at the end of the class. The two topics are:
- Predictive control & model-based reinforcement learning;
- Neural network theory: learning & generalisation.
(3-0-6) TTh 2:30-4:00.
- Yisong Yue (email@example.com);
- Ugo Rosolia (firstname.lastname@example.org);
- Jeremy Bernstein (email@example.com).
- Natalie Bernat (firstname.lastname@example.org);
- Joe Marino (email@example.com);
- Alex Farhang (firstname.lastname@example.org).
- Weeks 1–3: lectures on reinforcement learning & control;
- Weeks 4–6: lectures on neural network theory;
- Weeks 7–10: research project and special guest lectures.
Office hours will be held on the class Discord server. A signup link will be sent to students during the first week of class. Students are encouraged to use the Discord freely at other times to discuss homeworks, projects and any other class-related matters. If you do not have access to the Discord by the end of the first week of class, email an instructor.
It goes without saying, but please remember to treat other students with respect and don’t say anything you wouldn’t in a Caltech classroom.
The final grade will be based 20% on homeworks and 80% on the final project.
Students should complete homeworks in groups of two to three—submitting one solution document per group. All students in a group should understand all parts of the completed homework. If you don’t have a group, use the
#matchmaking channel on Discord to find one.
Completed homeworks should be submitted on Gradescope. If you do not have access to the Gradescope by the end of the first week of class, email an instructor.
Homeworks will be set on Thursday and due the Thursday after.
|#||Date set||Date due||Resources|
A group has a total of 48 late hours for the term. This means that if the first homework is submitted 47 hours late and the second homework is submitted two hours late, the group would incur no penalty on the first homework but score zero on the second.
The final project will be conducted during the last four weeks of class. Students should work in teams of two to three. Students may either pick one of the special topics to build their project on, or alternatively they may attempt to mix the two topics. The poster session is on 6/03 and project reports are due 6/04.
|0||3/30||Introduction||pdf / vid|
|Topic 1—RL & Control|
|1||3/30||Discrete MDPs||pdf / vid|
|2||4/01||Optimal Control||pdf / vid|
|3||4/06||Model Predictive Control||pdf / vid|
|4||4/08||Learning MPC||pdf / vid / supp|
|5||4/13||Model Learning in MPC||pdf / vid|
|6||4/15||Planning Under Uncertainty and Project Ideas||pdf / vid|
|Topic 2—Neural Network Theory|
|7||4/20||Neural Architecture Design||pdf / vid|
|8||4/22||Network Function Spaces||pdf / vid / ipynb|
|9||4/27||Network Optimisation||pdf / vid|
|10||4/29||Statistical Learning Theory||pdf / vid|
|11||5/04||PAC-Bayesian Theory||pdf / vid|
|12||5/06||Project Ideas||pdf / vid|
|Final Project & Guest Lectures|
|13||5/11||Joe Marino||pdf / vid|
|14||5/13||Yasaman Bahri||pdf / vid|
|16||5/20||Guanya Shi||pdf / vid|
|17||5/25||Roberto Calandra||pdf / vid|
|18||5/27||Guillermo Valle-Pérez & Ard Louis||pdf / vid|
|19||6/01||SueYeon Chung||pdf / vid|