This package "statisticalRL" encapusulates 3 PYPI packages:
- statisticalRL-environments: dedicated to definitions of environments such as Multi-armed Bandits and Makov Decision Processes
- statisticalRL-learners: dedicated to definitions of learning agents such as Upper Confidence Bound (UCB) agent or Upper Confidence for Reinforcement Learning (UCRL) agent.
- statisticalRL-experiments: dedicated to running experiments, storing and visualizing results to produce plots ready for integration in a research article.
On top of that, the project contains a repository named "article" to reproduce experiments from research articles.