On Filter Size in Graph Convolutional Networks

By Dinh Van Tran, Nicoló Navarin, Alessandro Sperduti. 2018

53 citations recorded by the source corpus.

Recently, many researchers have been focusing on the definition of neural networks for graphs. The basic component for many of these approaches remains the graph convolution idea proposed almost a decade ago. In this paper, we extend this basic component, following an intuition derived from the well-known convolutional filters over multi-dimensional tensors. In particular, we derive a simple, efficient and effective way to introduce a hyper-parameter on graph convolutions that influences the filter size, i.e., its receptive field over the considered graph. We show with experimental results on real-world graph datasets that the proposed graph convolutional filter improves the predictive performance of Deep Graph Convolutional Networks.

Subfields: Artificial Intelligence, Information Systems, Statistical and Nonlinear Physics

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Shirui Pan
2. Guodong Long
3. Chengqi Zhang
4. Philip S. Yu
5. Zonghan Wu
6. Fengwen Chen
7. Jing Jiang
8. Charu C. Aggarwal

In Information Systems

1. Chengqi Zhang
2. Guodong Long
3. Shirui Pan
4. Philip S. Yu
5. Yongfeng Zhang
6. Jiawei Han
7. B. Liu
8. Bing Liu

In Statistical and Nonlinear Physics

1. Shirui Pan
2. Guodong Long
3. Chengqi Zhang
4. Philip S. Yu
5. Zonghan Wu

Citations in the corpus

• A Comprehensive Survey on Graph Neural Networks (2019, background): “65 82 2 [26], [46], [52], [54], [96], [98] PROTEIN [123] 1113 39.”