AI RESEARCH
Riemannian Stochastic Optimization for Sufficient Dimension Reduction
arXiv CS.LG
•
ArXi:2606.00413v1 Announce Type: cross Sufficient dimension reduction (SDR) makes high-dimensional regression tractable by projecting the covariates onto a low-dimensional subspace that preserves the conditional mean of the response. Existing gradient-based estimators either operate in the ambient space and suffer from the curse of dimensionality, or localize in the reduced space at a per-outer-iteration cost at least quadratic in the sample size.