Implemented and experimented with reinforcement learning algorithms, including TD(0), SARSA, Q Learning, and Deep Q-Networks, to optimize agent performance in OpenAI’s CartPole environment. Compared policies to visualize the efficiency of Deep Reinforcement Learning over normal reinforcement learning algorithms.
Python, PyTorch, NumPy, Pandas, and Matplotlib
- TD(0) Policy Algorithm
- SARSA Policy Algorithm
- Q Learning Policy Algorithm
- Deep Q-Networks Policy Algorithm