vicgpt/reinforcement-learning

Implementation of multi arm bandits for various use cases

Overview

This repository is aimed to provide a peek into different reinforcement algorithms and their use case. The structure of the repository is around different case that's used in the industry and research to solve reinforcement learning problems. Each use case will have their own seperate folders with the required code, documentation, basic description of the algorithms and reference. The documentation provided within the jupyter notebook will not be sufficient to understand the the theoritical background of different approaches, throughout this work I have the following books that can provide detail understanding -

• Reinforcement Learning: An Introduction (2nd Edition)
• Introduction to Multi arm bandits
• A Tutorial on Thompson Sampling

Content

• Online experiments with Multi-Arm bandits