Saved in:
Bibliographic Details
Main Authors: Lu, Jielong, Wu, Zhihao, Cai, Zhiling, Pi, Yueyang, Wang, Shiping
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.13426
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913800908701696
author Lu, Jielong
Wu, Zhihao
Cai, Zhiling
Pi, Yueyang
Wang, Shiping
author_facet Lu, Jielong
Wu, Zhihao
Cai, Zhiling
Pi, Yueyang
Wang, Shiping
contents In recent years, Graph Convolutional Networks (GCNs) have gained popularity for their exceptional ability to process graph-structured data. Existing GCN-based approaches typically employ a shallow model architecture due to the over-smoothing phenomenon. Current approaches to mitigating over-smoothing primarily involve adding supplementary components to GCN architectures, such as residual connections and random edge-dropping strategies. However, these improvements toward deep GCNs have achieved only limited success. In this work, we analyze the intrinsic message passing mechanism of GCNs and identify a critical issue: messages originating from high-order neighbors must traverse through low-order neighbors to reach the target node. This repeated reliance on low-order neighbors leads to redundant information aggregation, a phenomenon we term over-aggregation. Our analysis demonstrates that over-aggregation not only introduces significant redundancy but also serves as the fundamental cause of over-smoothing in GCNs.
format Preprint
id arxiv_https___arxiv_org_abs_2504_13426
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Simplifying Graph Convolutional Networks with Redundancy-Free Neighbors
Lu, Jielong
Wu, Zhihao
Cai, Zhiling
Pi, Yueyang
Wang, Shiping
Machine Learning
In recent years, Graph Convolutional Networks (GCNs) have gained popularity for their exceptional ability to process graph-structured data. Existing GCN-based approaches typically employ a shallow model architecture due to the over-smoothing phenomenon. Current approaches to mitigating over-smoothing primarily involve adding supplementary components to GCN architectures, such as residual connections and random edge-dropping strategies. However, these improvements toward deep GCNs have achieved only limited success. In this work, we analyze the intrinsic message passing mechanism of GCNs and identify a critical issue: messages originating from high-order neighbors must traverse through low-order neighbors to reach the target node. This repeated reliance on low-order neighbors leads to redundant information aggregation, a phenomenon we term over-aggregation. Our analysis demonstrates that over-aggregation not only introduces significant redundancy but also serves as the fundamental cause of over-smoothing in GCNs.
title Simplifying Graph Convolutional Networks with Redundancy-Free Neighbors
topic Machine Learning
url https://arxiv.org/abs/2504.13426