Graph Signal Processing and Deep Learning: Convolution, Pooling, and Topology

被引：40

作者：

Cheung, Mark ^{[1
]}

Shi, John ^{[1
]}

Wright, Oren ^{[2
]}

Jiang, Lavendar Y. ^{[1
,3
]}

Liu, Xujin ^{[1
]}

Moura, Jose M. F. ^{[4
,5
,6
,7
]}

机构：

[1] Carnegie Mellon Univ, Elect & Comp Engn Dept, Pittsburgh, PA 15213 USA

[2] Carnegie Mellon Univ, Inst Software Engn, Pittsburgh, PA 15213 USA

[3] Carnegie Mellon Univ, Dept Math Sci, Pittsburgh, PA 15213 USA

[4] Carnegie Mellon Univ, Pittsburgh, PA 15213 USA

[5] Amer Assoc Advancement Sci, Pittsburgh, PA USA

[6] Portugal Acad Sci, Lisbon, Portugal

[7] US Natl Acad Engineers, Washington, DC USA

来源：

IEEE SIGNAL PROCESSING MAGAZINE | 2020年 / 37卷 / 06期

基金：

美国国家科学基金会; 美国安德鲁·梅隆基金会;

关键词：

Convolution; Deep learning; Computer architecture; Graphical models; Filtering theory; Discrete Fourier transforms;

D O I：

10.1109/MSP.2020.3014594

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Deep learning, particularly convolutional neural networks (CNNs), has yielded rapid, significant improvements in computer vision and related domains. But conventional deep learning architectures perform poorly when data have an underlying graph structure, as in social, biological, and many other domains. This article explores 1) how graph signal processing (GSP) can be used to extend CNN components to graphs to improve model performance and 2) how to design the graph CNN architecture based on the topology or structure of the data graph.

引用

页码：139 / 149

页数：11

共 30 条

[1]

[Anonymous], 2019, PROC ICLR WORKSHOP R

[2]

Bastian M., 2009, P INT AAAI C WEBL SO, P361, DOI 10.13140/2.1.1341.1520

[3] Geometric Deep Learning Going beyond Euclidean data [J].

Bronstein, Michael M. ;

Bruna, Joan ;

LeCun, Yann ;

Szlam, Arthur ;

Vandergheynst, Pierre .

IEEE SIGNAL PROCESSING MAGAZINE, 2017, 34 (04) :18-42

[4]

Bruna J., 2014, P INT C LEARN REPR I

[5]

Cheung M, 2019, CONF REC ASILOMAR C, P462, DOI [10.1109/ieeeconf44664.2019.9048796, 10.1109/IEEECONF44664.2019.9048796]

[6]

Defferrard M, 2016, ADV NEUR IN, V29

[7]

Du Jian, 2017, ARXIV171010370

[8]

Gao HY, 2019, PR MACH LEARN RES, V97

[9]

Hamilton WL, 2017, ADV NEUR IN, V30

[10]

Kipf TN, 2016, ICLR, DOI DOI 10.48550/ARXIV.1609.02907

← 1 2 3 →