Answers are grounded in this paper and cite their sources.
Methodological study demonstrates robust conditional mean estimation in functional linear regression, highlighting optimal convergence rates under heavy-tailed noise.
Robust estimation has played an important role in statistical and machine learning. However, its applications to functional linear regression are still underdeveloped. In this paper, we focus on Huber’s loss with a diverging robustness parameter which was previously used in parametric models. Compared to other robust methods such as median regression, the distinction is that the proposed method aims to estimate the conditional mean robustly, instead of estimating the conditional median. We only require [Formula: see text]th moment assumption ([Formula: see text]) on the noise distribution, and the established error bounds match the optimal rate in the least-squares case as soon as [Formula: see text]. We establish convergence rate in probability when the functional predictor has a finite 4th moment, and finite-sample bound with exponential tail when the functional predictor is Gaussian, in terms of both prediction error and [Formula: see text] error. The results also extend to the case of functional estimation in a reproducing kernel Hilbert space (RKHS).
No takes yet. Share an insight, caveat, or question.
Peng et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: