Research map / Learning theory and algorithms / Matrix and tensor methods

Generalized Laplacian in Spectral Seriation on Manifold Data

NeurIPS 2026 · Learning theory and algorithms › Matrix and tensor methods · approach: Low-rank approximation theory

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Studies spectral seriation for recovering latent circular phases from high-dimensional noisy observations on a manifold, proposing a generalized Laplacian variant of the standard spectral method.

Problem

Recovering latent phases t_i from high-dimensional observations Z_i with E[Z_i] = X(t_i), where X(t) parameterizes a manifold on the circle (e.g., estimating object-viewing angles from photographs).

Approach

Applies a generalized Laplacian to the similarity matrix in spectral seriation, extending the standard degree-normalized spectral method for phase recovery on manifold-structured data.

Experimental setup

Baselines: standard spectral seriation

Builds on: spectral seriation

Closest papers by meaning

• Nonlinear Laplacians: Tunable principal component analysis under directional prior information (NeurIPS 2025 · similarity 0.88)
• The Noisy Laplacian: a Threshold Phenomenon for Non-Linear Dimension Reduction (ICML 2025 · similarity 0.87)
• Coupled Laplacian Eigenmaps for Locally-Aware 3D Rigid Point Cloud Matching (CVPR 2024 · similarity 0.84)
• Wrapped Gaussian on the manifold of Symmetric Positive Definite Matrices (ICML 2025 · similarity 0.83)
• Spectral Perturbation Bounds for Low-Rank Approximation with Applications to Privacy (NeurIPS 2025 · similarity 0.83)
• CW Complex Hypothesis for Image Data (ICML 2024 · similarity 0.83)