Saved in:
Bibliographic Details
Main Authors: Han, Zhiwei, Matthes, Stefan, Shen, Hao
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.23785
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914356444266496
author Han, Zhiwei
Matthes, Stefan
Shen, Hao
author_facet Han, Zhiwei
Matthes, Stefan
Shen, Hao
contents We investigate the identifiability of nonlinear Canonical Correlation Analysis (CCA) in a multi-view setup, where each view is generated by an unknown nonlinear map applied to a linear mixture of shared latents and view-private noise. Rather than attempting exact unmixing, a problem proven to be ill-posed, we instead reframe multi-view CCA as a basis-invariant subspace identification problem. We prove that, under suitable latent priors and spectral separation conditions, multi-view CCA recovers the pairwise correlated signal subspaces up to view-wise orthogonal ambiguity. For $N \geq 3$ views, the objective provably isolates the jointly correlated subspaces shared across all views while eliminating view-private variations. We further establish finite-sample consistency guarantees by translating the concentration of empirical cross-covariances into explicit subspace error bounds via spectral perturbation theory. Experiments on synthetic and rendered image datasets validate our theoretical findings and confirm the necessity of the assumed conditions.
format Preprint
id arxiv_https___arxiv_org_abs_2602_23785
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Provable Subspace Identification of Nonlinear Multi-view CCA
Han, Zhiwei
Matthes, Stefan
Shen, Hao
Machine Learning
We investigate the identifiability of nonlinear Canonical Correlation Analysis (CCA) in a multi-view setup, where each view is generated by an unknown nonlinear map applied to a linear mixture of shared latents and view-private noise. Rather than attempting exact unmixing, a problem proven to be ill-posed, we instead reframe multi-view CCA as a basis-invariant subspace identification problem. We prove that, under suitable latent priors and spectral separation conditions, multi-view CCA recovers the pairwise correlated signal subspaces up to view-wise orthogonal ambiguity. For $N \geq 3$ views, the objective provably isolates the jointly correlated subspaces shared across all views while eliminating view-private variations. We further establish finite-sample consistency guarantees by translating the concentration of empirical cross-covariances into explicit subspace error bounds via spectral perturbation theory. Experiments on synthetic and rendered image datasets validate our theoretical findings and confirm the necessity of the assumed conditions.
title Provable Subspace Identification of Nonlinear Multi-view CCA
topic Machine Learning
url https://arxiv.org/abs/2602.23785