Machine Intelligence for Risk: Deep Learning Models in Credit, Fraud, and Market Prediction

Risk assessment has benefited enormously from the application of deep learning. Machine intelligence, natural-language processing, neural approximation, representation learning, and generative techniques are now routinely leveraged in the credit, fraud, and markets domains. Driven by academic curiosity, access to rich digital datasets, and some of the most powerful computing resources in history, an increasing number of researchers and practitioners are exploring deep learning for risk.