Exploring compact reinforcement-learning representations with linear regression

Open in new window