This repository contains a Jupyter Notebook detailing the development, training, and evaluation of a Reinforcement Learning (RL) agent on the CartPole-v1 environment from OpenAI Gym. The notebook employs Q-Learning, a model-free reinforcement learning algorithm, to teach an agent how to successfully balance a pole on a cart.
The solution outlined in the notebook discretizes the CartPole-v1 environment's continuous state space, enabling the implementation of Q-Learning. It guides through initializing the Q-table, defining the policy for action selection, updating Q-values based on rewards, and iterating this process to improve the agent's performance over time.
To run the code, open the ipynb file in your browser and click "Open in Colab".