{"id":"arxiv/2605.15082","url":"https://arxiv.org/abs/2605.15082","tags":["stat.ML","cs.LG","math.ST"],"score":0,"title":"Average Gradient Outer Product in kernel regression provably recovers the central subspace for multi-index models","domain":"stat.ML","source":"arxiv","authors":["Libin Zhu","Damek Davis","Dmitriy Drusvyatskiy","Maryam Fazel"],"reviews":0,"subject":"stat.ML","summary":"We study a prototypical situation when a learned predictor can discover useful low-dimensional structure in data, while using fewer samples than are needed for accurate prediction. Specifically, we consider the problem of recovering a multi-index polynomial $f^*(x)=h(Ux)$, with $U\\in\\mathbb{R}^{r\\times d}$ and $r\\ll d$, from finitely many data/label pairs. Importantly, the target function depends on input $x$ only through the projection onto an unknown $r$-dimensional central subspace. The algorithm we analyze is appealingly simple: fit kernel ridge regression (KRR) to the data and compute…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-05-14","reactions":{}}