Review of the topic of the contextual multi-armed bandit. Includes implementation of easy-to-extend building blocks that form the contextual-bandit problem - e.g. agent, oracle, policy, environment.
This repo contains a review of the contextual multi-armed bandits.
Includes proposed framework for extendible building blocks that form the contextual bandit problem.
You can find the overview of the contextual bandits, dataset, and the framework in the presentation