Abstract
Wide & Deep, a simple yet effective learning architecture for recommendation systems developed by Google, has had a significant impact in both academia and industry due to its com bination of the memorization ability of general ized linear models and the generalization ability of deep models. Graph convolutional net works (GCNs) remain dominant in node classification tasks; however, recent studies have high lighted issues such as heterophily and expressive ness, which focus on graph structure while seemingly neglecting the potential role of node features. In this paper, we propose a flexible frame work GCNIII, which leverages the Wide & Deep architecture and incorporates three techniques: Intersect memory, Initial residual and Identity mapping. We provide comprehensive empirical evidence showing that GCNIII can more effectively balance the trade-off between over-fitting and over-generalization on various semi- and full supervised tasks. Additionally, we explore the use of large language models (LLMs) for node feature engineering to enhance the performance of GCNIII in cross-domain node classification tasks. Our implementation is available at https: //github.com/CYCUCAS/GCNIII.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
"Wide & Deep Learning for Node Classification", Technical Disclosure Commons, (July 08, 2025)
https://www.tdcommons.org/dpubs_series/8324