Reinforcement learning was perhaps most famously used by Google DeepMind in 2016 to build AlphaGo, a program that learned for itself how to play the incredibly complex and subtle board game Go to ...