A car is trapped in a simulated slope. An Agent is able to act on the car by pushing and pulling on it.

Through many iterations, the Agent is able to learn to get the car out of the slope.

python 2.7
conda install matplotlib

conda create --name python27 python=2.7
source activate python27
conda install matplotlib

python learning-starter.py

This is an assignment from a class on Reinforcement Learning, taught by Richard Sutton, the father of Reinforcement Learning