Stationary signal processing on graphs
arXiv:1601.02522 · doi:10.1109/TSP.2017.2690388
Abstract
Graphs are a central tool in machine learning and information processing as they allow to conveniently capture the structure of complex datasets. In this context, it is of high importance to develop flexible models of signals defined over graphs or networks. In this paper, we generalize the traditional concept of wide sense stationarity to signals defined over the vertices of arbitrary weighted undirected graphs. We show that stationarity is expressed through the graph localization operator reminiscent of translation. We prove that stationary graph signals are characterized by a well-defined Power Spectral Density that can be efficiently estimated even for large graphs. We leverage this new concept to derive Wiener-type estimation procedures of noisy and partially observed signals and illustrate the performance of this new model for denoising and regression.
References in corpus (4)
Cited by in corpus (58)
- Connecting the Dots: Identifying Network Structure via Graph Signal Processing
- Stationary Graph Processes and Spectral Estimation
- Gated Graph Recurrent Neural Networks
- DeepSphere: Efficient spherical Convolutional Neural Network with HEALPix sampling for cosmological applications
- Graph Signal Processing: History, Development, Impact, and Outlook
- Greedy Sampling of Graph Signals
- Forecasting Time Series with VARMA Recursions on Graphs
- Fast Robust PCA on Graphs
- Advances in Distributed Graph Filtering
- Filtering Random Graph Processes Over Random Time-Varying Graphs
- Spectral Domain Sampling of Graph Signals
- Reconstruction of Time-varying Graph Signals via Sobolev Smoothness
- A Directed Graph Fourier Transform with Spread Frequency Components
- Graph Fourier Transform: A Stable Approximation
- Network Inference from Consensus Dynamics with Unknown Parameters
- Bayesian Estimation of Graph Signals
- Multivariate Relations Aggregation Learning in Social Networks
- Graph Signal Sampling Under Stochastic Priors
- Graph Signal Processing Meets Blind Source Separation
- The Issues of Mismodelling Gravitational-Wave Data for Parameter Estimation
- Gegenbauer Graph Neural Networks for Time-varying Signal Reconstruction
- Joint Inference of Multiple Graphs from Matrix Polynomials
- Graph Signal Processing: Vertex Multiplication
- Ergodicity in Stationary Graph Processes: A Weak Law of Large Numbers
- DeepSphere: towards an equivariant graph-based spherical CNN
- Joint Time-Vertex Fractional Fourier Transform
- Widely-Linear MMSE Estimation of Complex-Valued Graph Signals
- Joint Network Topology Inference in the Presence of Hidden Nodes
- Localized Fourier Analysis for Graph Signal Processing
- Multi-dimensional Graph Fourier Transform
- Graph Signal Processing -- Part II: Processing and Analyzing Signals on Graphs
- A Graph Signal Processing View on Functional Brain Imaging
- Reconstructing undirected graphs from eigenspaces
- Detecting Localized Categorical Attributes on Graphs
- Wide-Sense Stationarity in Generalized Graph Signal Processing
- Spatiotemporal Covariance Neural Networks
- Graph Signal Processing: Overview, Challenges and Applications
- Kernel-Based Models for Influence Maximization on Graphs based on Gaussian Process Variance Minimization
- From Spectrum Wavelet to Vertex Propagation: Graph Convolutional Networks Based on Taylor Approximation
- A Time-Vertex Signal Processing Framework
- Extreme Values of the Fiedler Vector on Trees
- Bayesian Design of Sampling Set for Bandlimited Graph Signals
- Mask Combination of Multi-layer Graphs for Global Structure Inference
- Design of Sampling Set for Bandlimited Graph Signal Estimation
- Adaptive Graph-based Total Variation for Tomographic Reconstructions
- Compressive PCA for Low-Rank Matrices on Graphs
- A Second Order Cumulant Spectrum Test That a Stochastic Process is Strictly Stationary and a Step Toward a Test for Graph Signal Strict Stationarity
- Signed Graph Learning with Hidden Nodes
- Sampling Policy Design for Tracking Time-Varying Graph Signals with Adaptive Budget Allocation
- Online non-parametric change-point detection for heterogeneous data streams observed over graph nodes
- Distributed interference cancellation in multi-agent scenarios
- Stationarity of Time-Series on Graph via Bivariate Translation Invariance
- Translation Operator in Graph Signal Processing: A Generalized Approach
- Offline detection of change-points in the mean for stationary graph signals
- Enhancing Geometric Deep Learning via Graph Filter Deconvolution
- Stationary Geometric Graphical Model Selection
- Estimation of Shortest Path Covariance Matrices
- Time-Varying Graph Learning with Constraints on Graph Temporal Variation