Covers: implementation of Reinforcement learning

Check this out: https://cs.stanford.edu/people/karpathy/reinforcejs/

You can see how a few different example RL algorithms work. I particularly like the Gridworld TD example - try changing the epsilon parameter and see how it effects learning. Also check out the Waterworld example. Write any interesting observations from this as a comment on this recipe!

Contributors

- Objectives
- This recipe provides a high level procedure for identifying and leveraging reinforcement learning related use cases in real world.
- Potential Use Cases
- stock trading, event scheduling
- Who is This For ?
- INTERMEDIATEanyone with high level understanding of machine learning and interested in potential RL use cases in industry

Click on each of the following **annotated items** to see details.

Resources3/13

VIDEO 1. RL in Real World

- How can RL be used in the real world?

60 minutes

PAPER 2. An empirical investigation of the challenges of real-world reinforcement learning

- What challenged does using RL in real world face?

10 minutes

CALL_TO_ACTION 3. Example RL agents learning in the browser

10 minutes

UPLOAD_PDF 4. RL in Real World

- How can RL be used in the real world?

20 minutes

VIDEO 5. Deep RL Class Video Playlist

- How does deep reinforcement learning work?

3 hours

OTHER 6. THE RL Book

- Where can i find everything about reinforcement learning?

15 minutes

OTHER 7. Udacity RL Class

- What is reinforcement learning?

10 minutes

OTHER 8. Coursera RL specialization from U Alberta

- What is reinforcement learning?
- How can I use reinforcement learning?

3 hours

ARTICLE 9. Spinning Up in Deep RL

- How can I quickly start implementing deep RL?

10 minutes

OTHER 10. Algorithms in Reinforcement Learning

- What algorithms are used in reinforcement learning?
- Why might you pick one algorithm over another?

10 minutes

ARTICLE 11. Markov Decision Process

- What is MDP?

5 minutes

ARTICLE 12. Game Theory

- What is game theory?
- How does game theory relate to RL?

5 minutes

ARTICLE 13. Using Dynamic Programming to find the optimal policy in Grid World

- What is value iteration?
- What is policy iteration?

10 minutes

This is interesting.