图分类《Hierarchical Graph Representation Learning with Differentiable Pooling》阅读笔记

图分类《Hierarchical Graph Representation Learning with Differentiable Pooling》阅读笔记

Abstract

对于graph neural networks (GNNs)，大体上分三种任务：Node Classification，Link Prediction和 Graph Classification。本文关注的是Graph Classification。作者提到，现有的GNN本质上是flat的，而没有学习到hierarchical的图表达。原文描述如下：

‘’However, current GNN methods are inherently flat and do not learn hierarchical representations of graphs—a limitation that is especially problematic for the task of graph classification, where the goal is to predict the label associated with an entire graph.‘’

为此，作者提出了DIFFPOOL，一种能够学习到图的层次表达并可以结合多种端到端的GNNs结构可微的图池化模块。DIFFPOOL为每一层的nodes学习可微的软聚类(soft cluster assignment)， 将nodes映射到一组簇中，这一组簇将作为下一层GNN Layer的输入(coarsened input)。

Introduction

问题分析

对于flat的不足和hierarchical的必要性，作者描述为：flat的本质是聚合信息只能通过edges完成而不能以hierarchical的方式推断和聚合信息。原文描述如下：

“A major limitation of current GNN arch