EUVIMEDEUVIMED entdecken
Crossref

DeloopSGNN: Revisiting Spectral GNNs Through the Lens of Spatial Aggregation

Duanyu Li, Huijun Wu, Min Xie, Kai Lu, Wenzhe Zhang, Zhenwei Wu, Yong Dong, Ruibo Wang

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (27) · 22878-22886

DOI: 10.1609/aaai.v40i27.39451 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Duanyu Li und 7 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Graph Neural Networks (GNNs) have been studied from two primary perspectives: spectral, which employs global graph signal filtering and is theoretically more expressive, and spatial, which builds on local neighborhood aggregation and generalizes well across diverse graph structures. While spectral GNNs are expected to perform better in theory, they often underperform in practice compared to spatial models. To better understand this gap, we introduce a novel theoretical framework for converting spectral GNNs into the spatial domain, allowing for more intuitive analysis. This transformation reveals that signal looping and repeated high-order aggregation are major causes of over-smoothing in spectral GNNs. By addressing these issues in the spatial domain and converting the model back to the spectral domain, we propose DeloopSGNN, a spectral GNN with improved expressive capacity. Experiments on benchmark datasets show that DeloopSGNN achieves consistently strong performance in terms of accuracy and adversarial robustness, demonstrating that spectral GNNs can benefit significantly from careful architectural design grounded in our proposed framework.

Zitieren

APA

Li, D., Wu, H., Xie, M., Lu, K., Zhang, W., Wu, Z., Dong, Y., & Wang, R. (2026). DeloopSGNN: Revisiting Spectral GNNs Through the Lens of Spatial Aggregation. Proceedings of the AAAI Conference on Artificial Intelligence, 40(27), 22878-22886. https://doi.org/10.1609/aaai.v40i27.39451

Vancouver

Li D, Wu H, Xie M, Lu K, Zhang W, Wu Z, et al. DeloopSGNN: Revisiting Spectral GNNs Through the Lens of Spatial Aggregation. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(27):22878-22886. doi:10.1609/aaai.v40i27.39451

BibTeX
@article{Li2026,
  author = {Li, Duanyu and Wu, Huijun and Xie, Min and Lu, Kai and Zhang, Wenzhe and Wu, Zhenwei and Dong, Yong and Wang, Ruibo},
  title = {DeloopSGNN: Revisiting Spectral GNNs Through the Lens of Spatial Aggregation},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {27},
  pages = {22878--22886},
  doi = {10.1609/aaai.v40i27.39451}
}
Kostenlos starten →