Residual Gated Graph ConvNets

Residual Gated Graph ConvNets
复制标题

DOI：
--
复制
发表时间：
2017-11
期刊：
ArXiv
影响因子：
--
通讯作者：
X. Bresson;T. Laurent
X. Bresson;T. Laurent
中科院分区：
其他
文献类型：
--
作者：
X. Bresson;T. Laurent

文献摘要

被引文献

相似文献

图结构数据，如社交网络、脑功能网络、基因调控网络、通信网络等，引起了人们对将深度学习技术推广到图域的兴趣。在本文中，我们感兴趣的是为可变长度的图设计神经网络，以解决诸如顶点分类、图分类、图回归和图生成任务等学习问题。大多数现有的工作都集中在循环神经网络（rnn）上，以学习图的有意义的表示，最近引入了新的卷积神经网络（ConvNets）。在这项工作中，我们想要严格比较这两个基本的体系结构家族来解决图学习任务。我们回顾了现有的图RNN和ConvNet体系结构，并提出了LSTM和ConvNet对任意大小图的自然扩展。然后，我们针对子图匹配和图聚类这两个基本图问题设计了一组分析控制实验来测试不同的架构。数值结果表明，本文提出的图卷积神经网络比图rnn准确率提高3-17%，速度提高1.5-4倍。图卷积神经网络也比变分（非学习）技术准确36%。最后，最有效的图卷积网络架构使用门控边和残差。残差在学习多层架构中起着至关重要的作用，因为它们提供了10%的性能增益。
Graph-structured data such as social networks, functional brain networks, gene regulatory networks, communications networks have brought the interest in generalizing deep learning techniques to graph domains. In this paper, we are interested to design neural networks for graphs with variable length in order to solve learning problems such as vertex classification, graph classification, graph regression, and graph generative tasks. Most existing works have focused on recurrent neural networks (RNNs) to learn meaningful representations of graphs, and more recently new convolutional neural networks (ConvNets) have been introduced. In this work, we want to compare rigorously these two fundamental families of architectures to solve graph learning tasks. We review existing graph RNN and ConvNet architectures, and propose natural extension of LSTM and ConvNet to graphs with arbitrary size. Then, we design a set of analytically controlled experiments on two basic graph problems, i.e. subgraph matching and graph clustering, to test the different architectures. Numerical results show that the proposed graph ConvNets are 3-17% more accurate and 1.5-4x faster than graph RNNs. Graph ConvNets are also 36% more accurate than variational (non-learning) techniques. Finally, the most effective graph ConvNet architecture uses gated edges and residuality. Residuality plays an essential role to learn multi-layer architectures as they provide a 10% gain of performance.