A Mixture-of-Experts Framework for Practical Hybrid-Quantum Models in Credit Card Fraud Detection

TL;DR

This study explores hybrid quantum-classical machine learning models for credit card fraud detection, demonstrating improved precision with minimal additional inference time by selectively routing transactions.

Contribution

It introduces a mixture-of-experts framework integrating quantum and classical models, showing practical benefits in fraud detection performance.

Findings

Hybrid models achieved higher average precision scores than classical models.

Selective routing to quantum-classical models reduces false positives.

Additional inference time is minimal, between 7 to 21 minutes.

Abstract

This paper investigates whether hybrid quantum-classical machine learning can deliver practical improvements in financial fraud detection performance for card-based and other payment transactions. Building on a Guided Quantum Compressor architecture, the approach integrates an autoencoder, a variational quantum circuit, and a classical neural head, and then embeds this hybrid model into a mixture-of-experts framework including a state-of-the-art gradient-boosted tree classifier. Using a European credit card dataset with severe class imbalance, the routed hybrid architecture with 0.6 threshold achieves average precision scores of $0.793\pm0.085$ compared to $0.770\pm0.096$ of XGBoost on 3 repeated 5-fold cross-validation benchmarks. Precision and recall comparisons reveals a possible trade-off of fraud and nominal detections with a reduction in false positives at the cost of a small reduction in fraud detections. The improvements are achieved while adding only 7 to 21 minutes of extra inference time depending on the choice of hyperparameters. These results indicate that selectively routing transactions to quantum-classical models can enhance fraud detection while remaining compatible with the latency and operational constraints of modern financial institutions.