...Repo for the Deep Reinforcement Learning Nanodegree program
The observation space consists of 8 variables corresponding to the position and velocity of the ball and racket. Each agent receives its own, local observation. Two continuous actions are available, correspondin