Grassmanian Interpolation of Low-Pass Graph Filters: Theory and Applications

2025/10/27 by Savostianov, Anton, Schaub, Michael T., Stamm, Benjamin
#FOS: Computer and information sciences #FOS: Electrical engineering #FOS: Mathematics #Machine Learning (cs.LG) #Numerical Analysis (math.NA) #Signal Processing (eess.SP) #Social and Information Networks (cs.SI) #Spectral Theory (math.SP) #electronic engineering #information engineering

paper · doi:10.48550/arxiv.2510.23235

Abstract

Low-pass graph filters are fundamental for signal processing on graphs and other non-Euclidean domains. However, the computation of such filters for parametric graph families can be prohibitively expensive as computation of the corresponding low-frequency subspaces, requires the repeated solution of an eigenvalue problem. We suggest a novel algorithm of low-pass graph filter interpolation based on Riemannian interpolation in normal coordinates on the Grassmann manifold. We derive an error bound estimate for the subspace interpolation and suggest two possible applications for induced parametric graph families. First, we argue that the temporal evolution of the node features may be translated to the evolving graph topology via a similarity correction to adjust the homophily degree of the network. Second, we suggest a dot product graph family induced by a given static graph which allows to infer improved message passing scheme for node classification facilitated by the filter interpolation.

Citations

• Maximum volume coordinates for Grassmann interpolation: Lagrange, Hermite, and errors
• A Quasi Time-Reversible scheme based on density matrix extrapolation on the Grassmann manifold for Born-Oppenheimer Molecular Dynamics
• A Survey on Spectral Graph Neural Networks
• Learning time-dependent PDE solver using Message Passing Graph Neural Networks
• Grassmann extrapolation of density matrices for Born-Oppenheimer molecular dynamics
• Signal processing on simplicial complexes
• Is Homophily a Necessity for Graph Neural Networks?
• Handling Missing Data with Graph Representation Learning
• Strong Consistency, Graph Laplacians, and the Stochastic Block Model
• Revisiting Graph Neural Networks: All We Have is Low-Pass Filters
• Manifold interpolation and model reduction
• Connecting the Dots: Identifying Network Structure via Graph Signal Processing
• Contextual Stochastic Block Models
• Graph Signal Processing: Overview, Challenges and Applications
• An introduction to Topological Data Analysis: fundamental and practical\n aspects for data scientists
• Geometric Deep Learning: Going beyond Euclidean data
• The emerging field of signal processing on graphs: Extending high-dimensional data analysis to networks and other irregular domains

Related

• Geometric Graph Filters and Neural Networks: Limit Properties and Discriminability Trade-offs
• GrassNet: State Space Model Meets Graph Neural Network
• Stability to Deformations of Manifold Filters and Manifold Neural Networks
• Manifold Graph Signal Restoration using Gradient Graph Laplacian Regularizer
• Robust Filter Design for Graph Signals