/k-bandit

Using a range of methods to approche the k armed bandit problem, inspired by Reinforcement Learning: An Introduction by Andrew Barto and Richard S. Sutton.

Primary LanguageJupyter NotebookMIT LicenseMIT

K-Bandit Problem

Following the examples from Reinforcement Learning: An Introduction by Andrew Barto and Richard S. Sutton.

To Do

  • Add non stationary k armed bandits
  • Fix Gibbs solver