Salvato in:
Dettagli Bibliografici
Autori principali: Su, Ye, Ye, Mingrui, Wang, Yining, Guo, Jipeng, Liu, Yong
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2604.10469
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914466486026240
author Su, Ye
Ye, Mingrui
Wang, Yining
Guo, Jipeng
Liu, Yong
author_facet Su, Ye
Ye, Mingrui
Wang, Yining
Guo, Jipeng
Liu, Yong
contents Standard resampling ratios (e.g., $α\approx 0.632$) are widely used as default baselines in ensemble learning for three decades. However, how these ratios interact with a base learner's intrinsic functional complexity in finite samples lacks a exact mathematical characterization. We leverage the Hoeffding-ANOVA decomposition to derive the first exact, finite-sample variance decomposition for subagging, applicable to any symmetric base learner without requiring asymptotic limits or smoothness assumptions. We establish that subagging operates as a deterministic low-pass spectral filter: it preserves low-order structural signals while attenuating $c$-th order interaction variance by a geometric factor approaching $α^c$. This decoupling reveals why default baselines often under-regularize high-capacity interpolators, which instead require smaller $α$ to exponentially suppress spurious high-order noise. To operationalize these insights, we propose a complexity-guided adaptive subsampling algorithm, empirically demonstrating that dynamically calibrating $α$ to the learner's complexity spectrum consistently improves generalization over static baselines.
format Preprint
id arxiv_https___arxiv_org_abs_2604_10469
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Exact Finite-Sample Variance Decomposition of Subagging: A Spectral Filtering Perspective
Su, Ye
Ye, Mingrui
Wang, Yining
Guo, Jipeng
Liu, Yong
Machine Learning
Standard resampling ratios (e.g., $α\approx 0.632$) are widely used as default baselines in ensemble learning for three decades. However, how these ratios interact with a base learner's intrinsic functional complexity in finite samples lacks a exact mathematical characterization. We leverage the Hoeffding-ANOVA decomposition to derive the first exact, finite-sample variance decomposition for subagging, applicable to any symmetric base learner without requiring asymptotic limits or smoothness assumptions. We establish that subagging operates as a deterministic low-pass spectral filter: it preserves low-order structural signals while attenuating $c$-th order interaction variance by a geometric factor approaching $α^c$. This decoupling reveals why default baselines often under-regularize high-capacity interpolators, which instead require smaller $α$ to exponentially suppress spurious high-order noise. To operationalize these insights, we propose a complexity-guided adaptive subsampling algorithm, empirically demonstrating that dynamically calibrating $α$ to the learner's complexity spectrum consistently improves generalization over static baselines.
title Exact Finite-Sample Variance Decomposition of Subagging: A Spectral Filtering Perspective
topic Machine Learning
url https://arxiv.org/abs/2604.10469