A Novel Higher-order Weisfeiler-Lehman Graph Convolution

C. Damke, V. Melnikov, E. Hüllermeier, in: S. Jialin Pan, M. Sugiyama (Eds.), Proceedings of The 12th Asian Conference on Machine Learning, PMLR, Bangkok, Thailand, 2020, pp. 49–64.

Download
OA 771.14 KB
Conference Paper | Published | English
Editor
;
Abstract
Current GNN architectures use a vertex neighborhood aggregation scheme, which limits their discriminative power to that of the 1-dimensional Weisfeiler-Lehman (WL) graph isomorphism test. Here, we propose a novel graph convolution operator that is based on the 2-dimensional WL test. We formally show that the resulting 2-WL-GNN architecture is more discriminative than existing GNN approaches. This theoretical result is complemented by experimental studies using synthetic and real data. On multiple common graph classification benchmarks, we demonstrate that the proposed model is competitive with state-of-the-art graph kernels and GNNs.
Publishing Year
Proceedings Title
Proceedings of The 12th Asian Conference on Machine Learning
Volume
129
Page
49-64
Conference
Asian Conference on Machine Learning
Conference Location
Bangkok, Thailand
Conference Date
2020-11-18 – 2020-11-20
LibreCat-ID

Cite this

Damke C, Melnikov V, Hüllermeier E. A Novel Higher-order Weisfeiler-Lehman Graph Convolution. In: Jialin Pan S, Sugiyama M, eds. Proceedings of The 12th Asian Conference on Machine Learning. Vol 129. Proceedings of Machine Learning Research. Bangkok, Thailand: PMLR; 2020:49-64.
Damke, C., Melnikov, V., & Hüllermeier, E. (2020). A Novel Higher-order Weisfeiler-Lehman Graph Convolution. In S. Jialin Pan & M. Sugiyama (Eds.), Proceedings of The 12th Asian Conference on Machine Learning (Vol. 129, pp. 49–64). Bangkok, Thailand: PMLR.
@inproceedings{Damke_Melnikov_Hüllermeier_2020, place={Bangkok, Thailand}, series={Proceedings of Machine Learning Research}, title={A Novel Higher-order Weisfeiler-Lehman Graph Convolution}, volume={129}, booktitle={Proceedings of The 12th Asian Conference on Machine Learning}, publisher={PMLR}, author={Damke, Clemens and Melnikov, Vitalik and Hüllermeier, Eyke}, editor={Jialin Pan, Sinno and Sugiyama, MasashiEditors}, year={2020}, pages={49–64}, collection={Proceedings of Machine Learning Research} }
Damke, Clemens, Vitalik Melnikov, and Eyke Hüllermeier. “A Novel Higher-Order Weisfeiler-Lehman Graph Convolution.” In Proceedings of The 12th Asian Conference on Machine Learning, edited by Sinno Jialin Pan and Masashi Sugiyama, 129:49–64. Proceedings of Machine Learning Research. Bangkok, Thailand: PMLR, 2020.
C. Damke, V. Melnikov, and E. Hüllermeier, “A Novel Higher-order Weisfeiler-Lehman Graph Convolution,” in Proceedings of The 12th Asian Conference on Machine Learning, Bangkok, Thailand, 2020, vol. 129, pp. 49–64.
Damke, Clemens, et al. “A Novel Higher-Order Weisfeiler-Lehman Graph Convolution.” Proceedings of The 12th Asian Conference on Machine Learning, edited by Sinno Jialin Pan and Masashi Sugiyama, vol. 129, PMLR, 2020, pp. 49–64.
All files available under the following license(s):
Creative Commons License:
CC-BYCreative Commons Attribution 4.0 International Public License (CC-BY 4.0)
Main File(s)
File Name
771.14 KB
Access Level
OA Open Access
Last Uploaded
2020-10-08T11:21:00Z

Supplementary Material
File Name
Access Level
OA Open Access
Last Uploaded
2020-10-08T11:24:29Z

Export

Marked Publications

Open Data LibreCat

Sources

arXiv 2007.00346

Search this title in

Google Scholar