Various reinforcement techniques like policy iteration, value iteration, Q-learning are studied and experimented on different OpenAI environments (using the gym python library). Also, the deep Q-Learning algorithms are used in the experiments to compare their performance with the traditional methods.
The code also focuses on the application of Deep Q-Learning on different OpenAI environments like CartPole, MsPacman, etc. Some additional experiments are performed in the case of CartPole to check how fast the DQN architecture converges.