RL: Resources

  • This course follows the 2nd-edition textbook by Richard Sutton and Andrew Barto, which can be downloaded for free here.
  • There is a new textbook on multi-agent RL by Stefano V. Albrecht, Filippos Christianos, Lukas Schäfer, which can be downloaded for free here. This book also contains introduction chapters on deep learning and deep RL.
  • For deep learning, there is a useful textbook by Ian Goodfellow, Yoshua Bengio and Aaron Courville which is available for free here. There is also a newer textbook by Simon Prince available for free here.
  • There is a useful textbook on bandit algorithms by Tor Lattimore and Csaba Szepesvari, which can be downloaded for free here.
  • See also this useful chapter on back-propagation in neural networks by Michael Nielsen.
  • Lecture videos from the Deep Learning and Reinforcement Learning Summer School 2018 in Toronto.
  • Useful paper on how to evaluate RL algorithms.
  • Survey papers on single-agent RL and multi-agent RL.
  • RL Algorithms Chart
License
All rights reserved The University of Edinburgh