Hierarchical Graph Representation Learning with Differentiable Pooling

06/22/2018
by   Rex Ying, et al.
2

Recently, graph neural networks (GNNs) have revolutionized the field of graph representation learning through effectively learned node embeddings, and achieved state-of-the-art results in tasks such as node classification and link prediction. 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. Here we propose DiffPool, a differentiable graph pooling module that can generate hierarchical representations of graphs and can be combined with various graph neural network architectures in an end-to-end fashion. DiffPool learns a differentiable soft cluster assignment for nodes at each layer of a deep GNN, mapping nodes to a set of clusters, which then form the coarsened input for the next GNN layer. Our experimental results show that combining existing GNN methods with DiffPool yields an average improvement of 5-10 benchmarks, compared to all existing pooling approaches, achieving a new state-of-the-art on four out of five benchmark data sets.

READ FULL TEXT
research
06/22/2018

Hierarchical Graph Representation Learning withDifferentiable Pooling

Recently, graph neural networks (GNNs) have revolutionized the field of ...
research
10/01/2020

Adaptive Multi-grained Graph Neural Networks

Graph Neural Networks (GNNs) have been increasingly deployed in a multit...
research
12/17/2019

Multi-Channel Graph Convolutional Networks

Graph neural networks (GNN) has been demonstrated to be effective in cla...
research
11/18/2019

ASAP: Adaptive Structure Aware Pooling for Learning Hierarchical Graph Representations

Graph Neural Networks (GNN) have been shown to work effectively for mode...
research
11/03/2018

Towards Sparse Hierarchical Graph Classifiers

Recent advances in representation learning on graphs, mainly leveraging ...
research
02/26/2019

EvolveGCN: Evolving Graph Convolutional Networks for Dynamic Graphs

Graph representation learning resurges as a trending research subject ow...
research
07/22/2020

Graph Neural Networks with Haar Transform-Based Convolution and Pooling: A Complete Guide

Graph Neural Networks (GNNs) have recently caught great attention and ac...

Please sign up or login with your details

Forgot password? Click here to reset