Proposed Credit Risk Model using Quantum Machine Learning

Abstract

In this study, we develop a binary classification credit risk model using quantum machine learning techniques. I propose a QSVM (Quantum Support-Vector Machine) in which the kernel function part is replaced by a quantum circuit, and construct a new credit risk model based on it. The results show that the proposed method outperforms the comparison method. This suggests that quantum machine learning is useful to some extent for constructing credit risk models.