Quantum features improve credit default prediction with linear models
Quantum Feature Engineering for Credit Default Prediction: When and Why IQP Circuits Help Linear Classifiers
Machine Learning
Summary
Predicting who will fail to repay credit is important because even small improvements can save money. The authors show that using special quantum circuits called IQP to create new features helps a simple linear model predict defaults better than classical methods, including a strong classical technique called Kernel PCA. They find the choice of which financial attributes to encode matters, and the quantum features work best when combined with these informative inputs. This reveals that quantum circuits can amplify signals in data rather than just invent new information.
What this means in practice
- •For financial risk teams: Incorporate quantum-generated features to enhance linear credit default models beyond classical nonlinear methods without increasing feature count.
- •For health insurance analysts: Use quantum feature maps from select clinical data attributes to improve simple predictive models of patient risk.
Tested on one dataset.
Authors
Menachem Finkelstein, Diana Legziel Levy, Zohar Yakhini, Sarel Cohen
Abstract
Credit default prediction is a tabular classification problem in which modest gains in F1 translate directly into reduced financial exposure. We ask whether Instantaneous Quantum Polynomial-time (IQP) circuits can produce features that improve a classifier over both its raw classical baseline and Kernel PCA - the strongest unsupervised classical non-linear alternative - at an equal feature budget. The dataset provides 23 financial attributes per client; for an n-qubit circuit we select n of them, encode each as a rotation angle, and read 2n expectation values back out as new features. The motivation for using a quantum circuit is computational: an n-qubit IQP circuit runs in constant depth and encodes feature correlations in a 2^n-dimensional Hilbert space, whereas classical simulation of its exact output statistics scales exponentially in n. Using the UCI Default of Credit Card Clients dataset and five-fold cross-validation, we find that appending 16 IQP features (n = 8 qubits) to a Logistic Regression model raises F1 from 0.462 to 0.517 (+0.055, p < 0.0001). Kernel PCA, the next-best method, reaches only 0.493 at the same feature count; the gap survives Benjamini-Hochberg correction across 12 tests (p = 0.00007). No other classifier - Random Forest, SVM, XGBoost, or k-NN - benefits, which points to a linear-expressivity mechanism rather than a generic improvement. We also show that how the 8 input features are chosen matters: Random Forest importance-guided selection reaches F1 = 0.523, while encoding maximally uncorrelated features drops it to 0.496, demonstrating that the circuit amplifies informative structure rather than creating it from scratch.