Fraud detection in credit card transactions using Quantum-Assisted Restricted Boltzmann Machines

TL;DR

This paper evaluates the effectiveness of quantum-assisted Restricted Boltzmann Machines in detecting credit card fraud, demonstrating superior performance over classical methods using real quantum hardware and large-scale transaction data.

Contribution

It introduces the application of quantum-assisted RBMs to real-world fraud detection, showing their potential advantages over classical approaches.

Findings

Quantum-assisted RBMs outperform classical methods in key metrics.

Current noisy quantum hardware can effectively implement these models.

The approach is scalable to large financial datasets.

Abstract

Use cases for emerging quantum computing platforms become economically relevant as the efficiency of processing and availability of quantum computers increase. We assess the performance of Restricted Boltzmann Machines (RBM) assisted by quantum computing, running on real quantum hardware and simulators, using a real dataset containing 145 million transactions provided by Stone, a leading Brazilian fintech, for credit card fraud detection. The results suggest that the quantum-assisted RBM method is able to achieve superior performance in most figures of merit in comparison to classical approaches, even using current noisy quantum annealers. Our study paves the way for implementing quantum-assisted RBMs for general fault detection in financial systems.