Graph connection Laplacian methods can be made robust to noise

DOI10.1214/14-AOS1275zbMATH Open1350.60036arXiv1405.6231OpenAlexW2230899220MaRDI QIDQ5963525FDOQ5963525

H.-T. Wu, Noureddine El Karoui

Publication date: 22 February 2016

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: Recently, several data analytic techniques based on connection graph laplacian (CGL) ideas have appeared in the literature. At this point, the properties of these methods are starting to be understood in the setting where the data is observed without noise. We study the impact of additive noise on these methods, and show that they are remarkably robust. As a by-product of our analysis, we propose modifications of the standard algorithms that increase their robustness to noise. We illustrate our results in numerical simulations.

Full work available at URL: https://arxiv.org/abs/1405.6231

Recommendations

• Graph connection Laplacian and random matrices with random blocks
• Perturbation of the eigenvectors of the graph Laplacian: application to image denoising
• Weighted graph Laplace operator under topological noise
• Learning Theory
• Towards a theoretical foundation for Laplacian-based manifold methods

zbMATH Keywords

noisekernel methodsrobustnessrandom matricesspectral geometryconcentration of measurevector diffusion mapsgraph connection Laplacian methods

Mathematics Subject Classification ID

Random matrices (probabilistic aspects) (60B20) Limit theorems in probability theory (60F99)

Cites Work

• Asymptotic Statistics
• Title not available (Why is that?)
• Title not available (Why is that?)
• Operator norm consistent estimation of large-dimensional sparse covariance matrices
• Asymptotics of graphical projection pursuit
• The concentration of measure phenomenon
• Geometric Representation of High Dimension, Low Sample Size Data
• Posterior convergence rates for Dirichlet mixtures of beta densities
• Vector diffusion maps and the connection Laplacian
• Title not available (Why is that?)
• Concentration of measure and spectra of random matrices: applications to correlation matrices, elliptical distributions and beyond
• Comparing Top k Lists
• Phase retrieval with polarization
• Eigenvector synchronization, graph rigidity and the molecule problem
• Ranking and Sparsifying a Connection Graph
• Viewing Angle Classification of Cryo-Electron Microscopy Images Using Eigenvectors
• Alternating projection, ptychographic imaging and phase synchronization
• On information plus noise kernel random matrices
• On robust regression with high-dimensional predictors
• Angular synchronization by eigenvectors and semidefinite programming
• Representation theoretic patterns in three dimensional cryo-electron microscopy. I: The intrinsic reconstitution algorithm
• A theoretical and computational framework for isometry invariant recognition of point cloud data
• A class of generalized Laplacians on vector bundles devoted to multi-channel image processing
• Title not available (Why is that?)
• Two-dimensional tomography from noisy projections taken at unknown random directions
• A Local Clustering Algorithm for Connection Graphs
• Continuous Procrustes Distance Between Two Surfaces
• A perfect match condition for point-set matching problems using the optimal mass transport approach

Cited In (17)

• Eigen-convergence of Gaussian kernelized graph Laplacian by manifold heat interpolation
• Explore Intrinsic Geometry of Sleep Dynamics and Predict Sleep Stage by Unsupervised Learning Techniques
• Spectral convergence of the connection Laplacian from random samples
• The geometry of synchronization problems and learning group actions
• The Steerable Graph Laplacian and its Application to Filtering Image Datasets
• The diffusion geometry of fibre bundles: horizontal diffusion maps
• Robust Inference of Manifold Density and Geometry by Doubly Stochastic Scaling
• Spatiotemporal analysis using Riemannian composition of diffusion operators
• Learning low-dimensional nonlinear structures from high-dimensional noisy data: an integral operator approach
• Embedding Riemannian manifolds by the heat kernel of the connection Laplacian
• Doubly Stochastic Normalization of the Gaussian Kernel Is Robust to Heteroskedastic Noise
• Solving Jigsaw Puzzles by the Graph Connection Laplacian
• Embeddings of Riemannian manifolds with finite eigenvector fields of connection Laplacian
• Approximations of the connection Laplacian spectra
• Think globally, fit locally under the manifold setup: asymptotic analysis of locally linear embedding
• Latent common manifold learning with alternating diffusion: analysis and applications
• Wave-shape oscillatory model for nonstationary periodic time series analysis

This page was built for publication: Graph connection Laplacian methods can be made robust to noise

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5963525)