How to Teach AI to Play Bell Non-Local Games: Reinforcement Learning

Motivated by the recent success of reinforcement learning in games such as Go and Dota2, we formulate Bell non-local games as a reinforcement learning problem. Such a formulation helps us to explore Bell non-locality in a range of scenarios. The measurement settings and the quantum states are selected by the learner randomly in the beginning. Still, eventually, it starts understanding the underlying patterns and discovers an optimal (or near-optimal) quantum configuration corresponding to the task at hand. We provide a proof of principle approach to learning quantum configurations for violating various Bell inequalities. The algorithm also works for non-convex optimization problems where convex methods fail, thus offering a possibility to explore optimal quantum configurations for Bell inequalities corresponding to large quantum networks. We also implement a hybrid quantum-classical variational algorithm with reinforcement learning.