Implementation of Reinforcement Learning algorithms in python.

Algorithms implemented:

1. epsilon-greedy on 10-armed bandit testbed
2. Softmax action selection method using the Gibbs distribution on a 10-armed testbed
3. UCB1
4. Median Elimination Algorithm
5. Q-learning on puddle-world using OpenAI Gym
6. SARSA on puddle-world using OpenAI Gym
7. SARSA-Lambda on puddle-world using OpenAI Gym
8. Policy Gradients on chakra & vishamC world using OpenAI Gym
9. SMDP Q-learning on four-room grid world environment using OpenAI Gym
10. Intra-Option Q-learning on four-room grid world environment using OpenAI Gym
11. Deep Q-Network (DQN) on ‘CartPole’ environment of OpenAI Gym using TensorFlow

Plots:

1. Regret
2. Average Reward
3. Percentage Optimal arm pulls
4. Visualizing Optimal Policy
5. Visualizing state values
6. The trajectory followed by learned agent
7. Learning Curves - Average steps to goal, Average total discounted return, Episode Length.