Id: arxiv/2611.20275 Url: https://arxiv.org/abs/2611.20275 Tags: stat.ML, cs.LG, math.ST Score: 0 Title: The Manifold-Index Model: provably and efficiently avoiding the statistical and computational curse of dimensionality Domain: stat.ML Source: arxiv Authors: Yantao Wu, Mauro Maggioni Reviews: 0 Subject: stat.ML Category: research Comments: 0 Disputed: false Division: stat Mentions: 0 Discussed: 0 Published: 2026-11-15 Page: https://delphi.science/r/arxiv/2611.20275 We study vector-valued regression with predictors filling a tube around an unknown $m$-dimensional manifold $\mathcal{M}$ in $\mathbb{R}^D$. In the Manifold-Index Model, the mean response is $F=f\circΠ_{\mathcal{M}}$, where $Π_{\mathcal{M}}$ is the nearest-point projection onto $\mathcal{M}$ and $f:\mathcal{M}\to\mathbb{R}^p$ is a possibly non-injective link function. Multiscale ANisotropic lOcal Regression (MANOR) estimates tangent planes by multiscale PCA of predictors with similar responses, then refines them by pooling outer products of local slopes. The neighborhoods narrow tangentially…