Graph signal processing tasks that leverage spectral information typically assume access to the complete graph topology, which is often unavailable in practice. We propose a systematic framework for subgraph filter learning (SFL), where subgraph-supported operators approximate ambient graph filters under partial observations.
arXiv:2607. 21263v1 Announce Type: new Abstract: Graph signal processing tasks that leverage spectral information typically assume access to the complete graph topology, which is often unavailable in practice.
arXiv:2602. 10031v2 Announce Type: replace Abstract: Graph neural networks (GNNs) are commonly divided into message-passing neural networks (MPNNs) and spectral GNNs, reflecting two largely separate research traditions in machine learning and signal processing.
arXiv:2609.05919v1 Announce Type: new Abstract: We propose a graph dictionary learning (GDL) framework where each graph is represented as a zero-mean Gaussian distribution derived from its filtered L...
The paper introduces Spectral Connectivity-Regularized Graph Learning (SCoGL), a method for learning sparse graphs from limited data by incorporating Laplacian spectral priors that promote global connectivity. SCoGL extends the graphical lasso objective with a connectivity prior derived from Laplacian eigenvalues and uses projected gradient descent with Armijo backtracking for optimization. Experiments demonstrate that SCoGL improves graph recovery and enhances downstream tasks such as graph signal denoising when observations are scarce.
The article reviews the problem of learning graph structures from data, noting that research has traditionally split into two paths: inferring the topology of a single graph from observations on it, and learning a generative distribution from multiple observed graphs to sample new ones. It proposes a unified framework that treats both as inverse problems of a common graph generation process, reviews key methods, and discusses their interrelations, strengths, and limitations. The review highlights opportunities for cross‑paradigm integration and outlines future research directions.
arXiv:2608. 12757v1 Announce Type: cross Abstract: Laplacian-regularized minimization is fundamental in signal processing and machine learning, but is limited by the dense and ill-conditioned nature of the graph Laplacian pseudoinverse.
The paper studies the stability of graph-aware continuous‑time generative models that use a graph filter combined with a learned graph neural network. It derives explicit Wasserstein bounds showing how relative graph perturbations affect the generated distributions, and proposes a principled framework for designing stable graph filters that preserve heat‑diffusion smoothing while improving structural stability. Experiments on synthetic and fMRI data demonstrate that these stable filters enhance robustness and match or surpass the generative quality of a heat‑equation baseline.
arXiv:2609.08152v1 Announce Type: new Abstract: Graph representation learning has largely focused on designing increasingly sophisticated models to transform graph topology into vector representation...
The paper demonstrates that high‑quality graph embeddings can be produced without complex models or training by propagating random features through topological structures derived from random walks and anonymous walks. These training‑free embeddings capture node proximity and structural roles, respectively, and perform competitively on node, edge, and graph tasks while often requiring less computation. Combining the two embedding types further improves inference quality for some tasks.
The paper presents provable guarantees for a spectral method that recovers binary node labels on signed graphs with edge‑flip noise. It provides graph‑structure‑agnostic bounds on approximate inference accuracy and maximum angle deviation, using matrix concentration and eigenvector perturbation techniques. The results connect to the Cheeger constant and are validated with synthetic experiments, marking the first theoretical analysis of this spectral approach.
The paper introduces a scalable subgraph sampling method that uses resistance curvature to guide the selection of nodes and edges for graph neural network training. It builds on ERC‑LG, a curvature approximation technique that employs Johnson‑Lindenstrauss projections and regularized multi‑GPU batched conjugate gradient solvers, thereby avoiding costly Laplacian pseudoinverse calculations and large embedding storage. Experiments demonstrate that ERC‑LG‑based sampling matches pseudoinverse‑based curvature numerically, runs faster than conjugate‑gradient‑only approaches, and achieves the best mean accuracy on six of seven real‑world node‑classification datasets.
arXiv:2607. 20857v1 Announce Type: cross Abstract: Scientific machine learning methods such as neural operators and physics-informed neural networks have advanced engineering applications and inverse problems, but their training typically requires large volumes of simulated data.