AI RESEARCH

Exact Finite-Sample Variance Decomposition of Subagging: A Spectral Filtering Perspective

arXiv CS.LG

ArXi:2604.10469v1 Announce Type: new Standard resampling ratios (e.g., $\alpha \approx 0.632$) 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.