Feature Ranking in Credit-Risk with Qudit-Based Networks
This work addresses the need for interpretable predictive models in finance, particularly for credit risk, but it is incremental as it applies a quantum method to a specific domain with competitive rather than groundbreaking results.
The authors tackled the problem of balancing accuracy and interpretability in credit-risk assessment by proposing a quantum neural network based on a single qudit, which outperformed logistic regression and matched random forest in macro-F1 score on a real-world dataset from Taiwan while providing interpretable feature rankings.
In finance, predictive models must balance accuracy and interpretability, particularly in credit risk assessment, where model decisions carry material consequences. We present a quantum neural network (QNN) based on a single qudit, in which both data features and trainable parameters are co-encoded within a unified unitary evolution generated by the full Lie algebra. This design explores the entire Hilbert space while enabling interpretability through the magnitudes of the learned coefficients. We benchmark our model on a real-world, imbalanced credit-risk dataset from Taiwan. The proposed QNN consistently outperforms LR and reaches the results of random forest models in macro-F1 score while preserving a transparent correspondence between learned parameters and input feature importance. To quantify the interpretability of the proposed model, we introduce two complementary metrics: (i) the edit distance between the model's feature ranking and that of LR, and (ii) a feature-poisoning test where selected features are replaced with noise. Results indicate that the proposed quantum model achieves competitive performance while offering a tractable path toward interpretable quantum learning.