Description: Welcome to the Reinforcement Learning Algorithms from Scratch repository! This project showcases a collection of essential Reinforcement Learning (RL) algorithms implemented entirely from scratch, without relying on any external RL libraries. Here, we delve into the fascinating world of RL, enabling you to gain a deeper understanding of the underlying concepts and mechanisms that power intelligent decision-making systems.

Motivation: The primary motivation behind this project is to demystify the inner workings of popular RL algorithms. While various RL libraries offer robust implementations, their complexity might obscure the fundamental principles for newcomers and enthusiasts alike. By crafting these algorithms from scratch, we aim to provide a transparent and educative journey that enables learners to grasp RL concepts more comprehensively.

Note:

Each jupyter notebook of this repository consists of the algorithms of one of the chapter of the following books:

Reinforcement Learning An Introduction Richard S. Sutton and Andrew G. Barto