https://medium.com/@monocosmo77/how-multi-agent-reinforcement-learning-is-developing-part1-artificial-intelligence-201a4ed5cfb8