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.
Comments
Accepted for presentation at IEEE High Performance Extreme Computing Conference (HPEC 2026)
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.
Banks predict credit default with Logistic Regression because regulators can read its coefficients—but it cannot express interactions. We add the missing non-linearity with a quantum-inspired feature map: 8 of the data's 23 columns become rotation angles of 8 qubits in an IQP circuit simulated in PennyLane, and 2 × 8 = 16 outputs are appended to the 23. On the UCI Default of Credit Card Clients data (30,000 clients, 5-fold CV) this lifts it from F1 = 0.462 to 0.517, while Kernel PCA, the strongest classical alternative at the same budget, reaches only 0.493; the gap survives FDR correction over 12 tests (p = 0.00007). The gain is specific to linear models: Random Forest, SVM, XGBoost, and k-NN, already non-linear, do not benefit. What caps n is the simulator, holding 2n amplitudes; on hardware the depth is independent of n.
Detecting credit card fraud is notoriously difficult because fraudulent charges make up less than 0.2% of total transactions. Quantum Machine Learning (QML), particularly Quantum Kernel Support Vector Machines (QSVM), promises richer feature representations by embedding classical data into high-dimensional Hilbert spaces. However, whether this theoretical advantage holds up when managing real-world financial data bottlenecks remains unclear. We designed a controlled benchmark comparing three quantum encodings, Angle, ZZFeatureMap, and IQP against classical models while keeping sample sizes, feature counts, and scaling strictly matched. To handle class skew without corrupting quantum state overlaps, we embedded synthetic minority oversampling (SMOTE) directly into the quantum Gram matrix workflow, raising PR-AUC scores by roughly +0.04 across feature maps. When trained on the full European credit card dataset with 30 features, classical baselines reached a top PR-AUC of 0.983. However, restricting classical models to match QSVM hardware constraints (500 samples, 4 PCA features) dropped their PR-AUC to 0.750. This puts our best quantum result 0.730 PR-AUC using IQP encoding on 150 samples into realistic context. Testing on physical quantum hardware reduced performance further to 0.62 due to CNOT gate errors and thermal noise. Overall, our findings pinpoint two main operational bottlenecks for near-term quantum kernels: the scaling of Gram matrix evaluation and heavy information loss from squeezing features onto small qubit registers.
Choosing which features to retain from a high-dimensional dataset is one of the more practically consequential decisions in building a machine learning pipeline, yet it is rarely treated as anything other than a preprocessing afterthought. In this paper we ask whether quantum computation can play a meaningful role in that decision. We present Variational Quantum Feature Selection (VQFS), a method that assigns trainable scalar weights to input features and folds those weights directly into the rotation angles of a parameterized quantum circuit. An Ll penalty on the weight vector encourages many weights to collapse toward zero during training, leaving a compact subset of features that the circuit has learned to rely on. Because the weights are differentiable, the whole system-feature selector and quantum classifier together-is trained end-to-end with gradient descent rather than through the combinatorial search that makes classical wrapper methods expensive. We built VQFS on top of the PennyLane simulator so that it runs on an ordinary laptop without access to quantum hardware. Testing on the Breast Cancer Wisconsin, Wine, and Iris benchmarks, we found that VQFS matched or beat five classical baselines (Lasso, PCA, Random Forest importance, Mutual Information, and RFE) on accuracy while cutting the feature count by 25-60%. Five-fold cross-validation showed the gains were consistent rather than the result of a lucky split. We see this work as a small but concrete step toward integrating quantum methods into the full machine learning workflow, not just the classification stage.
Feature selection for imbalanced classification tasks such as credit card fraud and consumer default detection requires balancing predictive relevance, inter-feature redundancy, and computational feasibility. We benchmark three computing paradigms, classical branch-and-bound optimization (Gurobi), photonic entropy computing (QCI Dirac-3), and simulated photonic boson sampling (Piquasso), across thirteen feature-selection methods on two datasets: ULB Credit Card Fraud (30 features) and AmEx consumer default (159 features). Each method is routed to the solver matched to its mathematical structure. On ULB, Dirac-3 MI-Spearman matches the all-features model using 13 of 30 features (mean F1 0.873 +/- 0.023 over five runs, best run 0.896), and Piquasso is the best method at k=5. On AmEx, performance rises steadily with the feature budget and every paradigm approaches F1 = 0.80 only near the full feature set. Most differences between Gurobi and Dirac-3 on identical methods fall within run-to-run variation; the large gaps occur where the certified optimum generalizes poorly, most sharply for distance correlation on AmEx at k=25 (Gurobi F1 = 0.422 vs. a Dirac-3 mean of 0.746). At matched budgets, F1 varies about ten times more across methods on ULB than on AmEx, which we trace to how concentrated the predictive signal is in each feature space.
Choosing which features to retain from a high-dimensional dataset is one of the more practically consequential decisions in building a machine learning pipeline, yet it is rarely treated as anything other than a preprocessing afterthought. In this paper we ask whether quantum computation can play a meaningful role in that decision. We present Variational Quantum Feature Selection (VQFS), a method that assigns trainable scalar weights to input features and folds those weights directly into the rotation angles of a parameterized quantum circuit. An Ll penalty on the weight vector encourages many weights to collapse toward zero during training, leaving a compact subset of features that the circuit has learned to rely on. Because the weights are differentiable, the whole system-feature selector and quantum classifier together-is trained end-to-end with gradient descent rather than through the combinatorial search that makes classical wrapper methods expensive. We built VQFS on top of the PennyLane simulator so that it runs on an ordinary laptop without access to quantum hardware. Testing on the Breast Cancer Wisconsin, Wine, and Iris benchmarks, we found that VQFS matched or beat five classical baselines (Lasso, PCA, Random Forest importance, Mutual Information, and RFE) on accuracy while cutting the feature count by 25-60%. Five-fold cross-validation showed the gains were consistent rather than the result of a lucky split. We see this work as a small but concrete step toward integrating quantum methods into the full machine learning workflow, not just the classification stage.
Financial fraud detection remains difficult due to extreme class imbalance and subtle, nonlinear patterns in transactional data. We investigate a hybrid quantum–classical approach that couples a fidelity-based quantum kernel with a logistic regression classifier (QEKLR). A custom ZZ-inspired feature map comprising Hadamards, mixed-axis Ry/Rzencodings, tunable Ising-ZZ cross terms, and data re-uploading, embeds PCA-reduced, Min–Max scaled features into quantum states. The resulting fidelity kernel is evaluated and used within kernel logistic regression. Experiments on two benchmarks: BankSim (synthetic) and the European Credit Card Fraud dataset, use balanced subsets created via Random Under-Sampling. The proposed model attains the best performance on BankSim (AUC = 0.96; strongest F1), and is competitive on the credit-card dataset (AUC = 0.94), closely tracking XGBoost (AUC = 0.96) while outperforming logistic regression and a variational quantum classifier baseline. These results indicate that quantum kernel methods can provide practical gains on tabular fraud tasks while remaining aligned with current NISQ constraints. We also discuss the quadratic cost of exact kernel evaluation and outline scalable approximations for future deployment on larger, highly imbalanced datasets.