Connection graph Laplacian methods can be made robust to noise

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.

33 pages

References in corpus (3)

• Consistency of spectral clustering
• Algorithms to automatically quantify the geometric similarity of anatomical surfaces
• Alternating Projection, Ptychographic Imaging and Phase Synchronization

Cited by in corpus (11)

• Single-lead f-wave extraction using diffusion geometry
• Latent common manifold learning with alternating diffusion: analysis and applications
• Embedding Riemannian Manifolds by the Heat Kernel of the Connection Laplacian
• Vector Nonlocal Euclidean Median: Principal Bundle Captures The Nature of Patch Space
• Wave-shape oscillatory model for nonstationary periodic time series analysis
• Embeddings of Riemannian Manifolds with Finite Eigenvector Fields of Connection Laplacian
• Diffuse to fuse EEG spectra -- intrinsic geometry of sleep dynamics for classification
• Think globally, fit locally under the Manifold Setup: Asymptotic Analysis of Locally Linear Embedding
• Efficient fetal-maternal ECG signal separation from two channel maternal abdominal ECG via diffusion-based channel selection
• Optimal Recovery of Precision Matrix for Mahalanobis Distance from High Dimensional Noisy Observations in Manifold Learning
• On the spectral property of kernel-based sensor fusion algorithms of high dimensional data