Credit Default Prediction with Projected Quantum Feature Models and Ensembles

TL;DR

This work tackles credit default prediction in large-scale, high-dimensional time-series data by proposing a hybrid quantum-classical framework that uses projected quantum features (PQF) and ensemble learning. The authors implement a near-term quantum approach featuring Heisenberg-based feature maps and PQFs, integrated with classical models like XGBoost, and evaluate on a public Default Prediction Dataset using a Composite Default Risk (CDR) metric. Across simulations and IBM hardware, ensemble models that combine quantum and classical predictions achieve modest yet consistent improvements over purely classical baselines, while demonstrating resilience to hardware noise. The results suggest practical viability for near-term quantum-enhanced ML in financial risk tasks and highlight open questions about scalable advantage and the potential for classical approximations of PQ features.

Abstract

Accurate prediction of future loan defaults is a critical capability for financial institutions that provide lines of credit. For institutions that issue and manage extensive loan volumes, even a slight improvement in default prediction precision can significantly enhance financial stability and regulatory adherence, resulting in better customer experience and satisfaction. Datasets associated with credit default prediction often exhibit temporal correlations and high dimensionality. These attributes can lead to accuracy degradation and performance issues when scaling classical predictive algorithms tailored for these datasets. Given these limitations, quantum algorithms, leveraging their innate ability to handle high-dimensionality problems, emerge as a promising new avenue alongside classical approaches. To assess the viability and effectiveness of quantum methodologies, we investigate a hybrid quantum-classical algorithm, utilizing a publicly available "Default Prediction Dataset" released as part of a third-party data science competition. Specifically, we employ hybrid quantum-classical machine learning models based on projected quantum feature maps and their ensemble integration with classical models to examine the problem of credit card default prediction. Our results indicate that the ensemble models based on the projected quantum features were capable of slightly improving the purely classical results expressed via a "Composite Default Risk" (CDR) metric. Furthermore, we discuss the practical applicability of the studied quantum-classical machine learning techniques and address open questions concerning their implementation.

Figures (4)

• Figure 1: Example of a quantum Heisenberg feature map circuit with $7$ qubits encoding $6$ features and $1$ repetition. $U_i$'s are single-qubit Haar-random unitaries.
• Figure 2: Diagram illustrating the projected quantum features (PQF) transformer. Note this can, for example be used in a pipeline, preceded by normalization and followed by a machine learning model.
• Figure 3: Topology of the quantum Heisenberg feature map circuit on ibm_marrakesh with 101 qubits encoding 174 features. Filled circles denote qubits, and connecting lines indicate couplings. Black circles and lines highlight the qubits and connections used in this experiment.
• Figure 4: Diversity between quantum and classical model predictions. Quantum model probabilities versus classical model probabilities across test dataset points.