Stacked SVD or SVD stacked? A Random Matrix Theory perspective on data integration
stat.ML, cs.LG, math.ST, stat.ME, stat.TH
Submitted: 2025-07-29
Updated: 2026-09-23
Comments: 61 pages. Analysis extended to handle multiple, partially shared components per dataset
Code: https://github.com/phillipnicol/stackedSVD
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Tracy-Widom, Gaussian, and Bootstrap: Approximations for Leading Eigenvalues in High-Dimensional PCA
- A Generalized Mean Approach for Distributed-PCA
- Entropic Optimal Transport Eigenmaps for Nonlinear Alignment and Joint Embedding of High-Dimensional Datasets
- Asymptotic limits of spiked eigenvalues and eigenvectors of signal-plus-noise matrices with weak signals and heteroskedastic noise
- Optimal Estimation of Shared Singular Subspaces across Multiple Noisy Matrices
- A spectral method for multi-view subspace learning using the product of projections
- Estimating shared subspace with AJIVE: the power and limitation of multiple data matrices
- Limit results for distributed estimation of invariant subspaces in multiple networks inference and PCA
Related papers
- Behavior of prediction performance metrics with rare events
- Optimal Estimation of Generic Dynamics by Path-Dependent Neural Jump ODEs
- A Posterior-Dynamics Framework for Imaging Inverse Problems with Pretrained Diffusion Priors
- Efficient quadratic entropy with distance sketches
- Embedding-Bias in Conditional Independence Testing
- From Log-Odds to Shapley Values: An Explanatory Geometry for the Weighted Naive Bayes Classifier