Graph signal processing for machine learning: A review and new perspectives
arXiv:2007.16061 · doi:10.1109/MSP.2020.3014591
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 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.
References in corpus (25)
- Modularity and community structure in networks
- Community detection in graphs
- 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
- A Review of Relational Machine Learning for Knowledge Graphs
- Discrete Signal Processing on Graphs
- Higher-order organization of complex networks
- Variational Graph Auto-Encoders
- Few-Shot Learning with Graph Neural Networks
- Learning graphs from data: A signal representation perspective
- Connecting the Dots: Identifying Network Structure via Graph Signal Processing
- Geometric Matrix Completion with Recurrent Multi-Graph Neural Networks
- Topological Signal Processing over Simplicial Complexes
- Stability Properties of Graph Neural Networks
- On the Convexity of Latent Social Network Inference
- Multitask learning over graphs: An Approach for Distributed, Streaming Machine Learning
- Filtering Random Graph Processes Over Random Time-Varying Graphs
- Compressive Spectral Clustering
- Multi-scale Modularity in Complex Networks
- Graph Learning from Data under Structural and Laplacian Constraints
- Relational Generalized Few-Shot Learning
- Understanding Graph Neural Networks from Graph Signal Denoising Perspectives
- Interpretable Neuron Structuring with Graph Spectral Regularization
- Modelling Graph Errors: Towards Robust Graph Signal Processing
- Robust Deep Graph Based Learning for Binary Classification
Cited by in corpus (20)
- Reconstruction of Time-varying Graph Signals via Sobolev Smoothness
- Graph Federated Learning for CIoT Devices in Smart Home Applications
- Joint Time-Vertex Fractional Fourier Transform
- Graph Normalized-LMP Algorithm for Signal Estimation Under Impulsive Noise
- Graph Signal Processing for Geometric Data and Beyond: Theory and Applications
- A Review of Graph-Powered Data Quality Applications for IoT Monitoring Sensor Networks
- Graph Neural Networks with Parallel Neighborhood Aggregations for Graph Classification
- Wide-Sense Stationarity in Generalized Graph Signal Processing
- Understanding Concepts in Graph Signal Processing for Neurophysiological Signal Analysis
- Interpretable Stability Bounds for Spectral Graph Filters
- A Comprehensive Benchmark on Spectral GNNs: The Impact on Efficiency, Memory, and Effectiveness
- Large Graph Signal Denoising with Application to Differential Privacy
- Graph Convolutional Neural Networks Sensitivity under Probabilistic Error Model
- On the Stability of Graph Convolutional Neural Networks under Edge Rewiring
- Combinatorial Complexes: Bridging the Gap Between Cell Complexes and Hypergraphs
- Learning Graph Filters for Structure-Function Coupling based Hub Node Identification
- How Powerful is Graph Convolution for Recommendation?
- Laplacian Constrained Precision Matrix Estimation: Existence and High Dimensional Consistency
- How likely is a random graph shift-enabled?
- Partition-wise Graph Filtering: A Unified Perspective Through the Lens of Graph Coarsening