README.md

A collection of implementations of the bandit problem (school project, Reinforcement Learning class, 2018).

Features

Bandits

• Normal multi-armed bandits
• Bernouilli multi-armed bandits

Algorithms

• Random bandit algorithm
• Greedy bandit algorithm
• Epsilon greedy bandit algorithm
• UCB bandit algorithm
• Thompson sampling algorithm

Using the code

See the notebook for examples of usage.

Credits

The notebook is from Valentin Emiya.