Seminar Advanced Deep Reinforcement Learning
Master Computer Science, Leiden University, The Netherlands
General information
The Seminar Advanced Deep Reinforcement Learning (SADRL) is a master level course taught at Leiden University.
The course is intended as a follow up to the master level courses Reinforcement Learning and Introduction to Deep Learning.
There are 12 places available (you will work in groups of 2).
Based on experience from previous years we advice you to (only) join when you scored an 8.0 or higher on both master courses Reinforcement Learning and Introduction to Deep Learning.
The goal of the course is to do actual research: re-run a state-of-the-art RL paper and/or even investigate a truly novel idea.
When you are lucky and your idea actually works you may be able to get a (workshop) publication from the course.
The course can also be an excellent starting point for an RL focused master's thesis afterwards.
Instructors: Aske Plaat & Thomas Moerland
Dates: September 12, 2022 - December 12, 2022 (every Monday)
Time: 11.00-13.00
Room: 403
Slides
Background
Reinforcement learning basics: The general basics of reinforcement learning can be studied in:
Book: Richard Sutton & Andrew Barto, Reinforcement Learning, an Introduction MIT Press 2018. http://incompleteideas.net/book/the-book.html
Deep reinforcement learning:
Book (studied in the master course Reinforcement Learning): Plaat, Aske. "Deep Reinforcement Learning." Springer. https://arxiv.org/pdf/2201.02135.pdf
Long Survey: François-Lavet, V., Henderson, P., Islam, R., Bellemare, M. G., & Pineau, J. (2018). An introduction to deep reinforcement learning. Foundations and Trends® in Machine Learning, 11(3-4), 219-354. https://arxiv.org/pdf/1811.12560.pdf
Short survey: Arulkumaran, K., Deisenroth, M. P., Brundage, M., & Bharath, A. A. (2017). Deep reinforcement learning: A brief survey. IEEE Signal Processing Magazine, 34(6), 26-38. https://discovery.ucl.ac.uk/id/eprint/10083557/1/1708.05866v2.pdf