Learning low-dimensional nonlinear structures from high-dimensional noisy data: an integral operator approach

DOI10.1214/23-aos2306arXiv2203.00126OpenAlexW4387828537MaRDI QIDQ6183757

Rong Ma, Xiucai Ding

Publication date: 4 January 2024

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

Full work available at URL: https://arxiv.org/abs/2203.00126

zbMATH Keywords

kernel methodspectral methodhigh-dimensional datamanifold learningnonlinear dimension reduction

Mathematics Subject Classification ID

Statistics on manifolds (62R30) Integral operators (47G10) Statistical aspects of big data and data science (62R07)

Cites Work

• On information plus noise kernel random matrices
• On Euclidean random matrices in high dimension
• The imbedding problem for Riemannian manifolds
• Geometry on probability spaces
• The spectrum of kernel random matrices
• Kernel methods in machine learning
• Data spectroscopy: eigenspaces of convolution operators and clustering
• The spectral norm of random inner-product kernel matrices
• On the distribution of the largest eigenvalue in principal components analysis
• Principal component analysis.
• Spectral convergence of graph Laplacian and heat kernel reconstruction in \(L^\infty\) from random samples
• Think globally, fit locally under the manifold setup: asymptotic analysis of locally linear embedding
• Optimality of spectral clustering in the Gaussian mixture model
• An \({\ell_p}\) theory of PCA and spectral clustering
• Statistical inference for principal components of spiked covariance matrices
• Error estimates for spectral convergence of the graph Laplacian on random geometric graphs toward the Laplace-Beltrami operator
• Statistical properties of kernel principal component analysis
• Consistency of spectral clustering
• Diffusion maps
• From graph to manifold Laplacian: the convergence rate
• Modern multidimensional scaling. Theory and applications.
• Learning theory estimates via integral operators and their approximations
• Singular vector and singular subspace distribution for the matrix denoising model
• Concentration of kernel matrices with application to kernel spectral clustering
• A Riemann-Stein kernel method
• THE SPECTRUM OF RANDOM KERNEL MATRICES: UNIVERSALITY RESULTS FOR ROUGH AND VARYING KERNELS
• THE SPECTRUM OF RANDOM INNER-PRODUCT KERNEL MATRICES
• Vector diffusion maps and the connection Laplacian
• Improving Spectral Clustering Using the Asymptotic Value of the Normalized Cut
• Empirical graph Laplacian approximation of Laplace–Beltrami operators: Large sample results
• An $\ell_{\infty}$ Eigenvector Perturbation Bound and Its Application to Robust Covariance Estimation
• Spectral convergence of the connection Laplacian from random samples
• Analysis of spectral clustering algorithms for community detection: the general bipartite setting
• High-Dimensional Probability
• Principal Manifolds and Nonlinear Dimensionality Reduction via Tangent Space Alignment
• Laplacian Eigenmaps for Dimensionality Reduction and Data Representation
• Spectral Convergence of Diffusion Maps: Improved Error Bounds and an Alternative Normalization
• Spectral Methods for Data Science: A Statistical Perspective
• Clustering with t-SNE, Provably
• Lipschitz Regularity of Graph Laplacians on Random Data Clouds
• On the Spectral Property of Kernel-Based Sensor Fusion Algorithms of High Dimensional Data
• Kernel Methods and Machine Learning
• The Dynamics of Message Passing on Dense Graphs, with Applications to Compressed Sensing
• Local Linear Regression on Manifolds and Its Geometric Interpretation
• Nonlinear Dimensionality Reduction
• Hessian eigenmaps: Locally linear embedding techniques for high-dimensional data
• DISTRIBUTION OF EIGENVALUES FOR SOME SETS OF RANDOM MATRICES
• Graph Based Gaussian Processes on Restricted Domains
• Scalability and robustness of spectral embedding: landmark diffusion is all you need
• Graph connection Laplacian methods can be made robust to noise
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item