Bayesian graph convolutional neural networks for semi-supervised classification
Yingxue Zhang Huawei Noah’s Ark Lab Montreal Research Centre 7101 Avenue du Parc, H3N 1X9 Montreal, QC Canada、Soumyasundar Pal∗ Mark Coates Dept. Electrical and Computer Engineering McGill University 3480 University St, H3A 0E9 Montreal, QC, Canada、Deniz Üstebay Huawei Noah’s Ark Lab Montreal Research Centre 7101 Avenue du Parc, H3N 1X9 Montreal, QC Canada These authors contributed equally to this work.
Abstract
Recently, techniques for applying convolutional neural networks to graph-structured data have emerged. Graph convolutional neural networks (GCNNs) have been used to address node and graph classification and matrix completion. Although the performance has been impressive, the current implementations have limited capability to incorporate uncertainty in the graph structure. Almost all GCNNs process a graph as though it is a ground-truth depiction of the relationship between nodes, but often the graphs employed in applications are themselves derived from noisy data or modelling assumptions. Spurious edges may be included; other edges may be missing between nodes that have very strong relationships. In this paper we adopt a Bayesian approach, viewing the observed graph as a realization from a parametric family of random graphs. We then target inference of the joint posterior of the random graph parameters and the node (or graph) labels. We present the Bayesian GCNN framework and develop an iterative learning procedure for the case of assortative mixed-membership stochastic block models. We present the results of experiments that demonstrate that the Bayesian formulation can provide bet
中文速览
图神经网络在处理图结构数据时，通常把输入的图当作节点关系的"完美真相"来使用，但现实中大多数图都是从含噪数据或模型假设中推导出来的，边的缺失或误加十分常见。这篇论文提出了一个贝叶斯图卷积神经网络（Bayesian GCNN）框架，把观测到的图视为某个参数化随机图模型的一次采样，再通过对随机图参数和网络权重做联合后验推断来完成节点分类，具体实现上采用了混合成员随机块模型（assortative mixed-membership stochastic block model）并结合 Monte Carlo dropout 进行近似推断。实验结果表明，在标注数据极为稀少的情况下，该贝叶斯方法的分类精度明显优于传统 GCNN，同时对图结构的随机扰动也表现出更强的鲁棒性。这项工作的意义在于，它首次将贝叶斯思想系统地引入图卷积网络，让模型能够显式地量化图结构不确定性，在标签匮乏或图质量较差的真实场景中具有重要的应用价值。
原文 arXiv:1811.11103;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1811.11103v1