Graph Signal Processing for Machine Learning: A Review and New Perspectives (2020-07-31T00:00:00.000000Z)

TL;DR

This article reviews a few important contributions made by GSP concepts and tools to the development of novel machine learning algorithms, and provides new perspectives on the future development of GSP techniques that may serve as a bridge between applied mathematics and signal processing and machine learning and network science.

Abstract

The effective representation, processing, analysis, and visualization of large-scale structured data, especially those related to complex domains, such as networks and graphs, are one of the key questions in modern machine learning. Graph signal processing (GSP), a vibrant branch of signal processing models and algorithms that aims at handling data supported on graphs, opens new paths of research to address this challenge. In this article, we review a few important contributions made by GSP concepts and tools, such as graph filters and transforms, to the development of novel machine learning algorithms. In particular, our discussion focuses on the following three aspects: exploiting data structure and relational priors, improving data and computational efficiency, and enhancing model interpretability. Furthermore, we provide new perspectives on the future development of GSP techniques that may serve as a bridge between applied mathematics and signal processing on one side and machine learning and network science on the other. Cross-fertilization across these different disciplines may help unlock the numerous challenges of complex data analysis in the modern age.

Authors

M. Bronstein

14 papers

Xiaowen Dong

2 papers

D. Thanou

1 papers

TL;DR

Abstract

Authors

References74 items

Understanding Graph Neural Networks from Graph Signal Denoising Perspectives

On The Stability of Polynomial Spectral Graph Filters

Multitask Learning Over Graphs: An Approach for Distributed, Streaming Machine Learning

Robust Deep Graph Based Learning for Binary Classification

Semi-Supervised Learning

Transferability of Spectral Graph Convolutional Neural Networks

Topological Signal Processing Over Simplicial Complexes

Relational Generalized Few-Shot Learning

Deep Unsupervised Learning of 3D Point Clouds via Graph Topology Inference and Filtering

Stability Properties of Graph Neural Networks

Introducing Graph Smoothness Loss for Training Deep Learning Architectures

Modelling Graph Errors: Towards Robust Graph Signal Processing

Zero Shot Learning with the Isoperimetric Loss

Approximating Spectral Clustering via Sampling: a Review

Bayesian graph convolutional neural networks for semi-supervised classification

STRING v11: protein–protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets

Connecting the Dots: Identifying Network Structure via Graph Signal Processing

Interpretable Neuron Structuring with Graph Spectral Regularization

Spectral MAB for Unknown Graph Processes

Relational inductive biases, deep learning, and graph networks

Learning Graphs From Data: A Signal Representation Perspective

Robust Graph Signal Processing in the Presence of Uncertainties on Graph Topology

Laplacian Power Networks: Bounding Indicator Function Smoothness for Adversarial Defense

Gaussian Processes Over Graphs

MOTIFNET: A MOTIF-BASED GRAPH CONVOLUTIONAL NETWORK FOR DIRECTED GRAPHS

An Inside Look at Deep Neural Networks Using Graph Signal Processing

Dynamic Graph CNN for Learning on Point Clouds

Graph Signal Processing: Overview, Challenges, and Applications

MARGIN: Uncovering Deep Neural Networks Using Graph Signal Analysis

Few-Shot Learning with Graph Neural Networks

Graph Attention Networks

Filtering Random Graph Processes Over Random Time-Varying Graphs

Geometric Matrix Completion with Recurrent Multi-Graph Neural Networks

Geometric Deep Learning on Graphs and Manifolds Using Mixture Model CNNs

Geometric Deep Learning: Going beyond Euclidean data

Variational Graph Auto-Encoders

Graph Learning from Data under Structural and Laplacian Constraints

Learning Heat Diffusion Graphs

Social Learning Over Weakly Connected Graphs

Learning sparse graphs under smoothness prior

Semi-Supervised Classification with Graph Convolutional Networks

Network Topology Inference from Spectral Templates

Higher-order organization of complex networks

Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering

Characterization and Inference of Graph Diffusion Processes From Observations of Stationary Signals

Compressive Spectral Clustering

How to Learn a Graph from Smooth Signals

Distributed Cooperative Optimal Control for Multiagent Systems on Directed Graphs: An Inverse Optimal Approach

A Review of Relational Machine Learning for Knowledge Graphs

Stability and continuity of centrality measures in weighted graphs

Big Data Analysis with Signal Processing on Graphs: Representation and processing of massive data sets with irregular structure

Graph Wavelets for Multiscale Community Mining

Learning Laplacian Matrix in Smooth Graph Signal Representations

Cooperative Optimal Control for Multi-Agent Systems on Directed Graph Topologies

Spectral Networks and Locally Connected Networks on Graphs

ImageNet classification with deep convolutional neural networks

The emerging field of signal processing on graphs: Extending high-dimensional data analysis to networks and other irregular domains

Discrete Signal Processing on Graphs

Kernels for Vector-Valued Functions: a Review

On the Convexity of Latent Social Network Inference

Opinion Dynamics and Learning in Social Networks

Inferring networks of diffusion and influence

Multi-scale modularity in complex networks

Wavelets on Graphs via Spectral Graph Theory

Community detection in graphs

Proto-value Functions: A Laplacian Framework for Learning Representation and Control in Markov Decision Processes

A tutorial on spectral clustering

Modularity and community structure in networks.

Sampling Techniques for Supervised or Unsupervised Tasks

A Comprehensive Survey on Graph Neural Networks

SP Authorized licensed use limited to: University of Oxford Libraries. Downloaded on October 31,2020 at 13:52:30 UTC from IEEE Xplore

Semi-supervised learning with graphs

Kernels and Regularization on Graphs

Model Selection Through Sparse Max Likelihood Estimation Model Selection Through Sparse Maximum Likelihood Estimation for Multivariate Gaussian or Binary Data

Field of Study

Journal Information