Master course in Reinforcement Learning (VUB, 2017-2018)

Welcome to the VUB course on reinforcement learning (RL), taught by Diederik M. Roijers and Ann Nowé. The TA for this course is Denis Steckelmacher.

All course materials will be placed on this website.

• The assignments until February are listed below
  
  
  • The MDP assignments (2 parts) are available here
  
  
  • The code for the MDP assigments is available here
  
  
  • The Multi-armed bandit assignments are available here
  
  


• Possible final projects (also watch your inboxes) can be found here
  
  


• The slides will be made available as the course progresses.
  
  
  • MDPs Introduction and Planning
  
  
  • RL: multi-armed bandits (with Denis Steckelmacher and Timothy Verstraeten)
  
  
  • RL: temporal difference methods
  
  
  • RL: Monte-Carlo methods
  
  
  • RL: model-based methods [video] (with Eugenio Bargiacchi)
  
  
  • Planning: partially observable MDPs
  
  
  • RL and Planning: multi-objective MDPs
  
  
  • RL: Hierarchical RL (by Denis Steckelmacher)