Projects I worked on while following the Deep Reinforcement Learning Nanodegree at Udacity
Implemented the Grid World environment and an agent using a Monte Carlo method. The project is under grid_world here
My solution for the OpenAI Gym's Taxi-v2 task, solved the task using Sarsa, SarsaMax, Expected Sarsa and Constant-alpha GLIE Monte Carlo Control. You can find the listing of the results I got with each algorithm under the project dir here
My solution for the first Udacity DRLND project 'Navigation' which consists of training an agent capable of collecting yellow bananas and avoiding blue ones while navigating in a square world. The project is under navigation here
My solution for the second Udacity DRLND project 'Continuous Control' which is about training a double-jointed arm to move to some target locations. The project can be found here
My solution for the third and final Udacity DRLND project 'Collaboration and Competition' which is about training two agents controlling rackets to bounce a ball over a net. The project can be found here