电子学报 ›› 2022, Vol. 50 ›› Issue (9): 2205-2214.DOI: 10.12263/DZXB.20210152

• 学术论文 • 上一篇    下一篇

基于剪边策略的图残差卷积深层网络模型

毛国君, 王者浩, 黄山, 王翔   

  1. 福建工程学院福建省大数据挖掘与应用重点实验室, 福建 福州 350118
  • 收稿日期:2021-01-24 修回日期:2021-12-31 出版日期:2022-09-25
    • 作者简介:
    • 毛国君 男,1966年月出生于内蒙古省赤峰市. 现为福建工程学院计算机与数学学院教授. 在国内外发表学术论文100余篇,知网引用3000以上.E-mail: 19662092@fjut.edu.cn
      王者浩 男,1995年3月出生于福建省福清市,现为福建工程学院在读研究生. 主要研究方向为电气工程人工智能.E-mail: wangzhehaofly@163.com
      黄 山 男,1987年5月出生于福建省武夷山市.现为福建工程学院计算机科学与数学学院讲师.研究方向为人工智能与深度学习、固体发光.E-mail: 19872162@fjut.edu.cn
      王 翔 男,1992年3月出生于福建省三明市.毕业于北京大学,获博士学位.现为福建工程学院计算机科学与数学学院副教授,硕士生导师.主要研究方向为人工智能,数字信号处理.
    • 基金资助:
    • 国家自然科学基金 (61773415)

A Deep Residual Graph Convolution Network Based on Dropedge Method

MAO Guo-jun, WANG Zhe-hao, HUANG Shan, WANG Xiang   

  1. Fujian Pvovincial Key Lab of Big Data Mining and Applications,Fujian University of Technology,Fuzhou,Fujian 350118,China
  • Received:2021-01-24 Revised:2021-12-31 Online:2022-09-25 Published:2022-10-26

摘要:

图神经网络自2005年以来已经逐步成为图学习中的一个重要的研究分支,其中最为活跃的是图卷积神经网络.由于图数据在现实世界中广泛存在,因此有效地完成图结构数据的学习具有很大的应用前景.目前出现的大多数图卷积神经网络模型基本都是浅层结构,过平滑问题成为制约该领域发展的瓶颈问题.本文提出了一种称为dri-GCN(Graph Convolutional Network via dropedge, residual and identity mapping)的图残差卷积深层网络模型,该模型集成了图剪边、初始残差和恒等映射技术.主要思想包括:利用图剪边技术增加学习数据的多样性,以防止学习过程中的过拟合现象;构建恒等映射下的初始残差网络,来扩展残差单元的学习路径,以削弱学习过程中的过平滑问题.实验结果表明,本文提出的dri-GCN模型可以帮助构建深层图卷积神经网络,通过网络层次的加深可以获得优于浅层网络的学习准确率.

关键词: 图神经网络, 图卷积神经网络, 剪边, 残差

Abstract:

Graph neural network (GNN) has become an important research field since 2005, and the most active branch is the graph convolutional network(GCN). In the real world, many applications are directly related to graphic data, so learning from graphic structure data is becoming more and more important, and can has a huge application prospect. However, most of the existing graph convolution neural network models can only support very few convoluted layers. This is mainly because of the over smoothing problem in GCNs, which has become one of the main bottlenecks in GCNs. In this paper, a novel GCN network: dri-GCN is proposed, which integrates the technologies of the Initial Residual, Identity Mapping and DropEdge into the traditional GCN. The main contributions include: using DropEdge to increase data diversity and prevent over-fitting; constructing the residual convolutional network under the constant mapping to extend the learning paths, which can effectively weaken the over smooth problem in GCNs. Experimental results show that dri-GCN model can help building the deeper graph convolution neural networks. There is no doubt that deeper GCNs can achieve better learning accuracy than shallower networks.

Key words: graph neural network, graph convolutional neural networks, DropEdge, residual network

中图分类号: