Simplifying Graph Convolutional Networks

19 Feb 2019Felix WuTianyi ZhangAmauri Holanda de Souza Jr.Christopher FiftyTao YuKilian Q. Weinberger

Graph Convolutional Networks (GCNs) and their variants have experienced significant attention and have become the de facto methods for learning graph representations. GCNs derive inspiration primarily from recent deep learning approaches, and as a result, may inherit unnecessary complexity and redundant computation... (read more)

PDF Abstract

Evaluation results from the paper


Task Dataset Model Metric name Metric value Global rank Uses extra
training data
Compare
Text Classification 20NEWS SGCN Accuracy 88.5 # 1
Sentiment Analysis MR SGCN Accuracy 75.9 # 11
Text Classification Ohsumed SGCN Accuracy 68.5 # 1
Text Classification R52 SGCN Accuracy 94.0 # 2
Text Classification R8 SGCN Accuracy 97.2 # 2
Relation Extraction TACRED C-SGC F1 67.0 # 8