Welcome to bayesianbandits’s documentation!

bayesianbandits is a Python package for Bayesian bandits. It provides a simple, Pythonic interface to implement and different algorithms to solve multi-armed bandit problems. To facilitate this, it provides a number of different Bayesian and recursive Bayesian estimators to represent the knowledge of the reward distribution of each arm.

If you have any questions or suggestions, please feel free to open an issue on the project page on GitHub or contact me at <rishi@kulkarni.science>.

Contents:

• Installation
• Usage and Examples
  • Notebooks
    • Thompson Sampling and the Bernoulli Bandit
    • Contextual Bandits for Count Data
    • A Bayesian Take on Linear Bandits and Disjoint LinUCB
    • Delayed and non-stationary multi-armed bandit problems
    • Bandit Persistence Recipes

API Reference

bayesianbandits.api	Agent API (bayesianbandits.api)
bayesianbandits	bayesianbandits (bayesianbandits)

Indices and tables

• Index

• Module Index

• Search Page