low-rank approximation 1matrix denoising 1minimax optimality 1multi-view learning 1singular subspace estimation 1
From the 1 of 2 linked papers with an AI index.
2 papers
math.ST2026
Optimal Estimation of Shared Singular Subspaces across Multiple Noisy Matrices
Zhengchi Ma, Rong Ma
The paper studies how to estimate common singular subspaces from several noisy low‑rank matrices, showing when a simple stacked SVD is optimal and proposing new methods for cases w…
stat.ML2025
Stacked SVD or SVD stacked? A Random Matrix Theory perspective on data integration
Tavor Z. Baharav, Phillip B. Nicol, Rafael A. Irizarry +1
Modern data analysis increasingly requires identifying shared latent structure across multiple high-dimensional datasets. A commonly used model assumes that the data matrices are n…