Graph Neural Networks in TensorFlow and Keras with Spektral

被引：113
作者：
Grattarola, Daniele [1 ]
Alippi, Cesare [1 ,2 ]
机构：
[1] Univ Svizzera Italiana, Lugano, Switzerland
[2] Politecn Milan, Milan, Italy
来源：
IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE | 2021年 / 16卷 / 01期
关键词：
Graph neural networks;
D O I：
10.1109/MCI.2020.3039072
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
Graph neural networks have enabled the application of deep learning to problems that can be described by graphs, which are found throughout the different fields of science, from physics to biology, natural language processing, telecommunications or medicine. In this paper we present Spektral, an open-source Python library for building graph neural networks with TensorFlow and the Keras application programming interface. Spektral implements a large set of methods for deep learning on graphs, including message-passing and pooling operators, as well as utilities for processing graphs and loading popular benchmark datasets. The purpose of this library is to provide the essential building blocks for creating graph neural networks, focusing on the guiding principles of user-fr iendliness and quick prototyping on which Keras is based. Spektral is, therefore, suitable for absolute beginners and expert deep learning practitioners alike. In this work, we present an overview of Spektral's features and report the performance of the methods implemented by the library in scenarios of node classification, graph classification, and graph regression.
引用
收藏
页码：99 / 106
页数：8
相关论文
共 53 条
[11]  
De Cao Nicola, 2018, ARXIV180809920
[12]  
Defferrard M, 2016, ADV NEUR IN, V29
[13]   Graph Transformation Policy Network for Chemical Reaction Prediction [J].
Do, Kien ;
Truyen Tran ;
Venkatesh, Svetha .
KDD'19: PROCEEDINGS OF THE 25TH ACM SIGKDD INTERNATIONAL CONFERENCCE ON KNOWLEDGE DISCOVERY AND DATA MINING, 2019, :750-760
[14]  
Du Jian, 2017, ARXIV171010370
[15]  
Farrell S., 2018, Novel deep learning methods for track reconstruction
[16]  
Fernandes Patrick, 2018, INT C LEARN REPR
[17]   Deciphering interaction fingerprints from protein molecular surfaces using geometric deep learning [J].
Gainza, P. ;
Sverrisson, F. ;
Monti, F. ;
Rodola, E. ;
Boscaini, D. ;
Bronstein, M. M. ;
Correia, B. E. .
NATURE METHODS, 2020, 17 (02) :184-+
[18]  
Gao H., 2019, Graph U-Nets
[19]  
Grattarola D., 2020, P 37 INT C MACH LEAR
[20]  
Gunnemann S., 2019, IINT C LEARN REPR IC
123456