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

Abstract

Deep learning, particularly convolutional neural networks (CNNs), have 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 paper explores 1)how graph signal processing (GSP) can be used to extend CNN components to graphs in order to improve model performance; and 2)how to design the graph CNN architecture based on the topology or structure of the data graph.

To be published on IEEE Signal Processing Magazine

References in corpus (13)

• A Comprehensive Survey on Graph Neural Networks
• The Emerging Field of Signal Processing on Graphs: Extending High-Dimensional Data Analysis to Networks and Other Irregular Domains
• Geometric deep learning: going beyond Euclidean data
• Spectral Networks and Locally Connected Networks on Graphs
• Discrete Signal Processing on Graphs
• Fast Graph Representation Learning with PyTorch Geometric
• Self-Attention Graph Pooling
• TUDataset: A collection of benchmark datasets for learning with graphs
• Topology Adaptive Graph Convolutional Networks
• Graph Neural Networks Exponentially Lose Expressive Power for Node Classification
• Graph Capsule Convolutional Neural Networks
• Graph Signal Processing: Modulation, Convolution, and Sampling
• Forecaster: A Graph Transformer for Forecasting Spatial and Time-Dependent Data

Cited by in corpus (5)

• Learning Dynamic Graph Representation of Brain Connectome with Spatio-Temporal Attention
• Decimated Framelet System on Graphs and Fast G-Framelet Transforms
• Graph Signal Processing: Vertex Multiplication
• Joint Time-Vertex Fractional Fourier Transform
• Explainable Machine Learning: The Importance of a System-Centric Perspective