Learning Augmentation for GNNs With Consistency Regularization

Graph neural networks (GNNs) have demonstrated superior performance in various tasks on graphs. However, existing GNNs often suffer from weak-generalization due to sparsely labeled datasets. Here we propose a novel framework that learns to augment the input features using topological information and...

Full description

Saved in:
Bibliographic Details
Published in:IEEE access 2021, Vol.9, p.127961-127972
Main Authors: Park, Hyeonjin, Lee, Seunghun, Hwang, Dasol, Jeong, Jisu, Kim, Kyung-Min, Ha, Jung-Woo, Kim, Hyunwoo J.
Format: Article
Language:English
Subjects:
Augmentation
Augmentors
Consistency
Data models
Datasets
Graph neural networks
Learning
meta-learning
Neural networks
Performance enhancement
Regularization
semi-supervised learning
Task analysis
Topology
Training
Training data
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Graph neural networks (GNNs) have demonstrated superior performance in various tasks on graphs. However, existing GNNs often suffer from weak-generalization due to sparsely labeled datasets. Here we propose a novel framework that learns to augment the input features using topological information and automatically controls the strength of augmentation. Our framework learns the augmentor to minimize GNNs' loss on unseen labeled data while maximizing the consistency of GNNs' predictions on unlabeled data. This can be formulated as a meta-learning problem and our framework alternately optimizes the augmentor and GNNs for a target task. Our extensive experiments demonstrate that the proposed framework is applicable to any GNNs and significantly improves the performance of graph neural networks on node classification. In particular, our method provides 5.78% improvement with Graph convolutional network (GCN) on average across five benchmark datasets.
ISSN:2169-3536
2169-3536
DOI:10.1109/ACCESS.2021.3111908