bmarroc/reinforcement-learning

Jupyter notebooks implementing Reinforcement Learning algorithms in Numpy and Tensorflow

Reinforcement Learning

This repository contains Jupyter notebooks implementing Reinforcement Learning algorithms to solve different problems.

The notebooks use Numpy 1.21.6 and Tensorflow 2.8.0.

Table of contents

• 1 - Option Pricing with Dynamic Programming
• 2 - Parking World with Monte Carlo
• 3 - Cliff Walking with TDL and SARSA
• 4 - Shortcut Maze with TDL and Q-Learning
• 5 - Random Walk with Function Approximation
• 6 - Mountain Car with Policy Gradient