Atari-Games-RL

A collection of ipython notebooks in which agents learn to play Atari games in Open AI gym environments using different methods of reinforcement learning.

Monte_Carlo.ipynb

Using monte carlo methods to find the optimal policy for the game "Blackjack"

Q-learning-cart.ipynb

Using deep Q-Learning to train an agent to play a game called "Cart and Pole"

Quadcopter_Project.ipynb

Designing and training an agent using Actor-Critic methods based on the DDPG algorithm to fly a quadcopter

Temporal_Difference.ipynb

Using temporal difference to find the optimal policy for the "CliffWalking" game