Litcius/Paper detail

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

Mark Cheung, John Shi, Oren Wright, Lavendar Y. Jiang, Xujin Liu, José M. F. Moura

2020IEEE Signal Processing Magazine64 citationsDOIOpen Access PDF

Abstract

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.

Topics & Concepts

Computer sciencePoolingDeep learningGraphArtificial intelligenceConvolutional neural networkTopological graph theoryTheoretical computer scienceSignal processingPattern recognition (psychology)Topology (electrical circuits)Machine learningVoltage graphDigital signal processingMathematicsLine graphCombinatoricsComputer hardwareAdvanced Graph Neural NetworksComplex Network Analysis TechniquesGraph Theory and Algorithms
Graph Signal Processing and Deep Learning: Convolution, Pooling, and Topology | Litcius