Reinforcment Learning Final Project

This is the final (mini) project, in the course Topics In Reinforcment learning, made by Meir Yagayev and Bar Siman Tov. In the course we solved 3 problems from OpenAI's Gym:

1 - Black Jack

An introduction assignment, using a table.

2 - Cart Pole

Solved using the SARSA algorithm, using all the features from the observation, as well as the accelerations(calculated as the difference between the last 2 velocities).

3 - Lunar Landing

This is the main project. Solved using Deep Learning, withe Keras and Tensorflow backend.