RL-Intro

This repo contains slides and code for a 30-minute ``drink from the firehose'' intro to RL, given at the OpenAI Hackthon on March 3rd, 2018.

The file pg_cartpole.py contains an 80-line, bare-bones implementation of policy gradient which seems to work in CartPole.

The file dqn_cartpole.py contains a short but somewhat messy implementation of DQN, which was developed during a Deep Q-Learning workshope at the Hackathon. Cheers to the participants of the workshop, who successfully debugged it!