Label-Enhanced Graph Neural Network for Semi-supervised Node Classification

31 May 2022  ·  Le Yu, Leilei Sun, Bowen Du, Tongyu Zhu, Weifeng Lv ·

Graph Neural Networks (GNNs) have been widely applied in the semi-supervised node classification task, where a key point lies in how to sufficiently leverage the limited but valuable label information. Most of the classical GNNs solely use the known labels for computing the classification loss at the output. In recent years, several methods have been designed to additionally utilize the labels at the input. One part of the methods augment the node features via concatenating or adding them with the one-hot encodings of labels, while other methods optimize the graph structure by assuming neighboring nodes tend to have the same label. To bring into full play the rich information of labels, in this paper, we present a label-enhanced learning framework for GNNs, which first models each label as a virtual center for intra-class nodes and then jointly learns the representations of both nodes and labels. Our approach could not only smooth the representations of nodes belonging to the same class, but also explicitly encode the label semantics into the learning process of GNNs. Moreover, a training node selection technique is provided to eliminate the potential label leakage issue and guarantee the model generalization ability. Finally, an adaptive self-training strategy is proposed to iteratively enlarge the training set with more reliable pseudo labels and distinguish the importance of each pseudo-labeled node during the model training process. Experimental results on both real-world and synthetic datasets demonstrate our approach can not only consistently outperform the state-of-the-arts, but also effectively smooth the representations of intra-class nodes.

PDF Abstract

Datasets


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Node Property Prediction ogbn-arxiv LEGNN + AS-Train Test Accuracy 0.7371 ± 0.0011 # 37
Validation Accuracy 0.7494 ± 0.0008 # 36
Number of params 5374120 # 10
Ext. data No # 1
Node Property Prediction ogbn-arxiv LEGNN Test Accuracy 0.7337 ± 0.0007 # 43
Validation Accuracy 0.7480 ± 0.0009 # 42
Number of params 5374120 # 10
Ext. data No # 1
Node Property Prediction ogbn-mag LEGNN + AS-Train Test Accuracy 0.5378 ± 0.0016 # 18
Validation Accuracy 0.5528 ± 0.0013 # 18
Number of params 5147997 # 28
Ext. data No # 1
Node Property Prediction ogbn-mag LEGNN Test Accuracy 0.5276 ± 0.0014 # 19
Validation Accuracy 0.5443 ± 0.0009 # 19
Number of params 5147997 # 28
Ext. data No # 1

Methods


No methods listed for this paper. Add relevant methods here