Papers › Improving Graph Neural Networks with Simple Architecture Design

Improving Graph Neural Networks with Simple Architecture Design

17 May 2021arXiv:2105.07634archive 2025-07-28

Sunil Kumar Maurya, Xin Liu, Tsuyoshi Murata

Graph Neural Networks have emerged as a useful tool to learn on the data by applying additional constraints based on the graph structure. These graphs are often created with assumed intrinsic relations between the entities. In recent years, there have been tremendous improvements in the architecture design, pushing the performance up in various prediction tasks. In general, these neural architectures combine layer depth and node feature aggregation steps. This makes it challenging to analyze the importance of features at various hops and the expressiveness of the neural network layers. As different graph datasets show varying levels of homophily and heterophily in features and class label distribution, it becomes essential to understand which features are important for the prediction tasks without any prior information. In this work, we decouple the node feature aggregation step and depth of graph neural network and introduce several key design strategies for graph neural networks. More specifically, we propose to use softmax as a regularizer and "Soft-Selector" of features aggregated from neighbors at different hop distances; and "Hop-Normalization" over GNN layers. Combining these techniques, we present a simple and shallow model, Feature Selection Graph Neural Network (FSGNN), and show empirically that the proposed model outperforms other state of the art GNN models and achieves up to 64% improvements in accuracy on node classification tasks. Moreover, analyzing the learned soft-selection parameters of the model provides a simple way to study the importance of features in the prediction tasks. Finally, we demonstrate with experiments that the model is scalable for large graphs with millions of nodes and billions of edges.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

sunilkmaurya/FSGNN officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Graph Neural NetworkNode Classificationfeature selection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Node Classification Actor FSGNN (8-hop) Accuracy 35.75 ± 0.96 #45 of 62 Archive leaderboard report
Node Classification Chameleon FSGNN (8-hop) Accuracy 78.27±1.28 #5 of 61 Archive leaderboard report
Node Classification Chameleon FSGNN (3-hop) Accuracy 78.14±1.25 #6 of 61 Archive leaderboard report
Node Classification Cornell FSGNN (8-hop) Accuracy 87.84±6.19 #3 of 60 Archive leaderboard report
Node Classification Squirrel FSGNN (8-hop) Accuracy 74.10±1.89 #5 of 59 Archive leaderboard report
Node Classification Texas FSGNN Accuracy 87.30 ± 5.55 #15 of 62 Archive leaderboard report
Node Classification Wisconsin FSGNN (3-hop) Accuracy 88.43±3.22 #13 of 63 Archive leaderboard report
Node Property Prediction ogbn-papers100M FSGNN Ext. data No #8 of 20 Archive leaderboard report
Node Property Prediction ogbn-papers100M FSGNN Number of params 16453301 #8 of 20 Archive leaderboard report
Node Property Prediction ogbn-papers100M FSGNN Test Accuracy 0.6807 ± 0.0006 #8 of 20 Archive leaderboard report
Node Property Prediction ogbn-papers100M FSGNN Validation Accuracy 0.7175 ± 0.0007 #8 of 20 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Feature SelectionGraph Neural NetworkSoftmax

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections