dennybritz/reinforcement-learning ? reverse-engineered prompt

Build me a reinforcement learning learning repo in Python that teaches the main ideas step by step and includes runnable notebook examples for each topic. I want clear sections for introduction, MDPs, dynamic programming, Monte Carlo methods, temporal difference learning, function approximation, deep Q learning, and policy gradient methods.

Each section should have a short explanation, learning goals, and worked exercises or solutions that show how the algorithms behave in simple OpenAI Gym environments. For the harder parts, use TensorFlow for the neural network examples, especially the DQN style notebooks. Include a few classic algorithms like policy evaluation, policy iteration, value iteration, Monte Carlo prediction and control, SARSA, Q learning, and at least one actor critic example.

Make it feel like a study companion to Sutton and Barto and David Silver’s course, with code that is easy to read and run in notebooks. If you need current library details, look up the latest docs online.