{"id":"arxiv/2604.01470","url":"https://arxiv.org/abs/2604.01470","tags":["math.ST","stat.ME"],"score":0,"title":"Sharp Debiasing for Smooth Functional Estimation in Banach Spaces","domain":"math.ST","source":"arxiv","authors":["Woonyoung Chang","Arun Kumar Kuchibhotla"],"reviews":0,"subject":"math.ST","summary":"This paper studies the estimation of smooth functionals $f(θ)$ of a mean parameter $θ= \\mathbb{E}_P[W]$ for a distribution $P$ on a general Banach space. We propose a cross-fitted estimator based on a single sample splitting and establish non-asymptotic moment bounds and Berry--Esséen bounds for both $m$-smooth and infinitely smooth functionals under the finite moment assumptions. Our framework is applied to precision matrix estimation and the inference of projection parameters in high-dimensional regression. In these Euclidean settings, the proposed estimators achieve asymptotic normality…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-04-01","reactions":{}}