Fraud detection in banking using deep reinforcement learning

Deep learning and machine learning are hot topics in the financial services nowadays. They allow financial entities to define products and segment clients, efficiently manage risk and detect fraud in banks. The theory of Deep Reinforcement Learning (DRL) was originally motivated by animal learning of sequential behavior, but has been developed and extended in the field of machine learning as an approach to Markov decision processes. Recently, a number of financial risk analysis and fraud detection studies have suggested a relationship between reward-related activities in the brain and functions necessary for DRL. Regarding the history of DRL, we introduce in this article the theory of DRL and present two applications in banking. Then we will discuss possible implementations.