Instructor: Daniel Hennes
Secretary:  Carola Stahl
Lectures: Tuesday, 17:30 – 19:00 (V38.03)
Tutorials: Wednesday, 14:00 – 15:30 (0.108) and Wednesday,  15:45 – 17:15 (0.447)
Office hours: by appointment
Communication: Announcements, such as change of schedule, will be communicated here and via the mailing list. Please subscribe to the mailing list to receive updates!

Lectures

Slides and references to the additional material will be posted as the course proceeds. Last years material is available here. Most slides adapted from material by Richard S. Sutton and previous years.

• Wed. April 11: Admin
• Tue. April 17:
  • Introduction
  • Multi-armed bandits
• Tue. April 24:
  • Finite Markov Decision Processes
  • room V38.01!
• Tue. May 1: no lecture
• Tue. May 8: Dynamic Programming
• Tue. May 15: Monte Carlo Methods
• Tue. May 22: no lecture
• Tue. May 29: Temporal-Difference Learning
• Tue. June 5: no lecture
• Tue. June 12:
  • n-step Bootstrapping
  • Planning and Learning
  • Additional reading: Multi-Step Reinforcement Learning: A Unifying Algorithm
• Tue. June 19: Value-function Approximation
• Tue. June 26: Policy Gradients
• Tue. July 3: Multi-agent Reinforcement Learning
• Tue. July 10: no lecture
• Tue. July 17: Case studies (canceled due to illness)
• Wed. July 18: Review + exam preparation (14:00 – 15:30 (0.108) , 15:45 – 17:15 (0.447))
• Wed. July 25: Exam 11:00 – 13:00 in V38.01 (tbc!)

Tutorials

Tutorials take place on Wednesdays, from 14:00 – 15:30 in room 0.108 and from 15:45 – 17:15 in room 0.447. At the beginning of each tutorial, you should sign the attendance list and mark which exercises you have successfully completed. We will select students at random to present their solutions. You need to complete at least 50% of the exercises to be allowed to the final exam.

• Wed. April 25: Exercise 01 is due.
• Wed. May 2
• Wed. May 9: Exercise 02 is due.
  • Solution to Question 5
• Wed. May 16: Exercise 03 is due.
• Wed. May 23
• Wed. May 30: Exercise 04 is due.
• Wed. June 6:
• Wed. June 13: Exercise 05 is due.
• Wed. June 20: Exercise 06 is due.
• Wed. June 27
• Wed. July 4: Exercise 07 is due.
• Wed. July 11
• Wed. July 18 (see above, review + exam prep) Prep. Exercise 08

Relevant textbooks

• R. Sutton & A. Barto: Reinforcement Learning: An Introduction (2nd edition) (pdf)
• C. Szepesvari: Algorithms for Reinforcement Learning (pdf)
• S. M. LaValle: Planning Algorithms (pdf)
• J. Norris: Markov Chains