David-Silver-Reinforcement-learning

This repository contains the notes for the Reinforcement Learning course by David Silver along with the implementation of the various algorithms discussed, both in Keras (with TensorFlow backend) and OpenAI's gym framework.

Syllabus:

• Week 1: Introduction to Reinforcement Learning [slide][video]

• Week 2: Markov Decision Processes [slide][video]

• Week 3: Planning by Dynamic Programming [slide][video]

• Week 4: Model-Free Prediction [slide][video]

• Week 5: Model-Free Control [slide][video]

• Week 6: Value Function Approximation [slide][video]

• Week 7: Policy Gradient Methods [slide][video]

• Week 8: Integrating Learning and Planning [slide][video]

• Week 9: Exploration and Exploitation [slide][video]

• Week 10: Case Study: RL in Classic Games [slide][video]

Dependencies

• TensorFlow
• Keras
• Gym
• Numpy

Install them using pip.

Contributing

Please feel free to create a Pull Request for adding implementations of the algorithms discussed in different frameworks like PyTorch, Caffe, etc. or improving the existing implementations. If you are a beginner, you can refer this for getting started.

Support

If you found this useful, please consider starring(★) the repo so that it can reach a broader audience.

License

This project is licensed under the MIT License - see the LICENSE file for details.

References

• https://github.com/dennybritz/reinforcement-learning
• https://github.com/llSourcell/AI_for_Video_Games_Syllabus