FraudGNAM: Inherently Interpretable Spectral GNN for Graph Fraud Detection

Suan Lee, Jinho Kim
Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence, PMLR 337:3398-3428, 2026.

Abstract

Graph neural networks for fraud detection face a fundamental tension: spectral methods that capture heterophilic patterns sacrifice interpretability, while interpretable models lack the spectral capacity to detect sophisticated fraud. We introduce {FraudGNAM} (Fraud Graph Neural Additive Model), the first inherently interpretable spectral GNN with learnable multi-scale filters that resolves this tension. {FraudGNAM} decomposes the fraud score into individually auditable components—per-feature contributions via neural additive models, pairwise feature interactions, multi-scale spectral band responses via learnable polynomial filters, and explicit feature-spectrum interactions—all combined through a transparent scalar-sum architecture. Each component’s contribution to the final prediction is independently quantifiable, unlike post-hoc explanation methods that approximate black-box decisions. On seven GADBench datasets spanning homophily ratios from 0.60 to 0.98, {FraudGNAM} achieves state-of-the-art AUROC on five of seven datasets against SEC-GFD, the strongest black-box baseline, while winning 14 of 21 metric-dataset comparisons overall. An adaptive configuration mechanism selects spectral order, high-pass filter count, and contrastive regularization based on measurable graph properties, eliminating per-dataset hyperparameter search.

Cite this Paper

BibTeX
@InProceedings{pmlr-v337-lee26e, title = {{FraudGNAM}: Inherently Interpretable Spectral GNN for Graph Fraud Detection}, author = {Lee, Suan and Kim, Jinho}, booktitle = {Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence}, pages = {3398--3428}, year = {2026}, editor = {Perković, Emilija and Malinsky, Daniel}, volume = {337}, series = {Proceedings of Machine Learning Research}, month = {17--21 Aug}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v337/main/assets/lee26e/lee26e.pdf}, url = {https://proceedings.mlr.press/v337/lee26e.html}, abstract = {Graph neural networks for fraud detection face a fundamental tension: spectral methods that capture heterophilic patterns sacrifice interpretability, while interpretable models lack the spectral capacity to detect sophisticated fraud. We introduce {FraudGNAM} (Fraud Graph Neural Additive Model), the first inherently interpretable spectral GNN with learnable multi-scale filters that resolves this tension. {FraudGNAM} decomposes the fraud score into individually auditable components—per-feature contributions via neural additive models, pairwise feature interactions, multi-scale spectral band responses via learnable polynomial filters, and explicit feature-spectrum interactions—all combined through a transparent scalar-sum architecture. Each component’s contribution to the final prediction is independently quantifiable, unlike post-hoc explanation methods that approximate black-box decisions. On seven GADBench datasets spanning homophily ratios from 0.60 to 0.98, {FraudGNAM} achieves state-of-the-art AUROC on five of seven datasets against SEC-GFD, the strongest black-box baseline, while winning 14 of 21 metric-dataset comparisons overall. An adaptive configuration mechanism selects spectral order, high-pass filter count, and contrastive regularization based on measurable graph properties, eliminating per-dataset hyperparameter search.} }
Endnote
%0 Conference Paper %T FraudGNAM: Inherently Interpretable Spectral GNN for Graph Fraud Detection %A Suan Lee %A Jinho Kim %B Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2026 %E Emilija Perković %E Daniel Malinsky %F pmlr-v337-lee26e %I PMLR %P 3398--3428 %U https://proceedings.mlr.press/v337/lee26e.html %V 337 %X Graph neural networks for fraud detection face a fundamental tension: spectral methods that capture heterophilic patterns sacrifice interpretability, while interpretable models lack the spectral capacity to detect sophisticated fraud. We introduce {FraudGNAM} (Fraud Graph Neural Additive Model), the first inherently interpretable spectral GNN with learnable multi-scale filters that resolves this tension. {FraudGNAM} decomposes the fraud score into individually auditable components—per-feature contributions via neural additive models, pairwise feature interactions, multi-scale spectral band responses via learnable polynomial filters, and explicit feature-spectrum interactions—all combined through a transparent scalar-sum architecture. Each component’s contribution to the final prediction is independently quantifiable, unlike post-hoc explanation methods that approximate black-box decisions. On seven GADBench datasets spanning homophily ratios from 0.60 to 0.98, {FraudGNAM} achieves state-of-the-art AUROC on five of seven datasets against SEC-GFD, the strongest black-box baseline, while winning 14 of 21 metric-dataset comparisons overall. An adaptive configuration mechanism selects spectral order, high-pass filter count, and contrastive regularization based on measurable graph properties, eliminating per-dataset hyperparameter search.
APA
Lee, S. & Kim, J.. (2026). FraudGNAM: Inherently Interpretable Spectral GNN for Graph Fraud Detection. Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 337:3398-3428 Available from https://proceedings.mlr.press/v337/lee26e.html.

Related Material