English

SStaGCN: Simplified stacking based graph convolutional networks

Machine Learning 2024-12-12 v2 Machine Learning

Abstract

Graph convolutional network (GCN) is a powerful model studied broadly in various graph structural data learning tasks. However, to mitigate the over-smoothing phenomenon, and deal with heterogeneous graph structural data, the design of GCN model remains a crucial issue to be investigated. In this paper, we propose a novel GCN called SStaGCN (Simplified stacking based GCN) by utilizing the ideas of stacking and aggregation, which is an adaptive general framework for tackling heterogeneous graph data. Specifically, we first use the base models of stacking to extract the node features of a graph. Subsequently, aggregation methods such as mean, attention and voting techniques are employed to further enhance the ability of node features extraction. Thereafter, the node features are considered as inputs and fed into vanilla GCN model. Furthermore, theoretical generalization bound analysis of the proposed model is explicitly given. Extensive experiments on 33 public citation networks and another 33 heterogeneous tabular data demonstrate the effectiveness and efficiency of the proposed approach over state-of-the-art GCNs. Notably, the proposed SStaGCN can efficiently mitigate the over-smoothing problem of GCN.

Keywords

Cite

@article{arxiv.2111.08228,
  title  = {SStaGCN: Simplified stacking based graph convolutional networks},
  author = {Jia Cai and Zhilong Xiong and Shaogao Lv},
  journal= {arXiv preprint arXiv:2111.08228},
  year   = {2024}
}
R2 v1 2026-06-24T07:39:59.097Z