Train a Smartcab How to Drive

Reinforcement Learning Project v2 2016/04/27 v1 2016/04/25

Instructor notes

Updated Smartcab.pdf - elaborated on the final discussion about optimal policy in respect to the agent.

FYI - Slightly modified environment.py and planner.py to notify the agent the game has finished by propagating the message to agent.LearningAgent.on_finished.

Files:

Report: Smartcab.pdf Report Source: https://docs.google.com/document/d/1Xfzw4powfsGocTZ8ANuXM4ahrjCKpCuyfB9jLfRFW8g/edit?usp=sharing Notepad for plots: smartcab_viz.ipynb

Install

This project requires Python 2.7 with the pygame library installed:

https://www.pygame.org/wiki/GettingStarted

Code

Open smartcab/agent.py and implement LearningAgent. Follow TODOs for further instructions.

Run

Make sure you are in the top-level project directory smartcab/ (that contains this README). Then run:

python smartcab/agent.py

OR:

python -m smartcab.agent

References

https://en.wikipedia.org/wiki/Q-learning http://artint.info/html/ArtInt_265.html