/Reinforcement-Learning-Problems

In this repository I try to solve small Reinforcement Learning problems using NumPy

Primary LanguagePythonMIT LicenseMIT

Reinforcement-Learning-Problems

In this repository I try to solve small Reinforcement Learning problems using NumPy. Current problems solved:

  1. Gridworld (State Value Estimation Problem)
  2. Gambler's Problem (State Value Estimation Problem)
  3. Windy Gridworld (Control Problem)
  4. K-Bandits (Control Problem)