In this project, I trained an agent to navigate (and collect bananas!) in a vast, square world. I used the Unity Machine Learning Agents Toolkit to design, train, and evaluate my own set of deep reinforcement learning algorithms implementations. I implemented Deep Q-learning (DQN), Double DQN (DDQN) and DDQN with prioritized experience replay. The environment used for this project is the Udacity version of the Banana Collector environment, from Unity. The goal of the agent is to collect as many yellow bananas as possible while avoiding blue bananas.

This project is connected to the Deep Reinforcement Learning Nanodegree, from Udacity. I used Python, Unity ML toolkit and Pytorch.