Saved in:
Bibliographic Details
Main Authors: Qian, Jian, Ge, Shu
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.18866
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914553402490880
author Qian, Jian
Ge, Shu
author_facet Qian, Jian
Ge, Shu
contents Despite the widespread use of boosting in structured prediction, a general theoretical understanding of aggregation beyond scalar prediction remains incomplete. We study vector-valued prediction under a target divergence and identify a geometric stability property under which aggregation amplifies weak guarantees into strong ones. We formalize this property as $(α,β)$-stability by geometric median and show how it supports a boosting framework based on exponential reweighting and geometric-median aggregation. For vector-valued prediction, we characterize this stability property under several natural divergences: $\ell_1$ and $\ell_2$ distances for unconstrained vector-valued prediction, and TV, Hellinger, and KL for density estimation over finite probability vectors. Building on these results, we propose a generic boosting framework \geomedboost. Under a weak learner condition and $(α,β)$-stability, we obtain exponential decay of the empirical divergence error, which then yields population guarantees through a generalization bound.
format Preprint
id arxiv_https___arxiv_org_abs_2602_18866
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle $(α,β)$-Stability for Boosting Vector-Valued Prediction
Qian, Jian
Ge, Shu
Machine Learning
Despite the widespread use of boosting in structured prediction, a general theoretical understanding of aggregation beyond scalar prediction remains incomplete. We study vector-valued prediction under a target divergence and identify a geometric stability property under which aggregation amplifies weak guarantees into strong ones. We formalize this property as $(α,β)$-stability by geometric median and show how it supports a boosting framework based on exponential reweighting and geometric-median aggregation. For vector-valued prediction, we characterize this stability property under several natural divergences: $\ell_1$ and $\ell_2$ distances for unconstrained vector-valued prediction, and TV, Hellinger, and KL for density estimation over finite probability vectors. Building on these results, we propose a generic boosting framework \geomedboost. Under a weak learner condition and $(α,β)$-stability, we obtain exponential decay of the empirical divergence error, which then yields population guarantees through a generalization bound.
title $(α,β)$-Stability for Boosting Vector-Valued Prediction
topic Machine Learning
url https://arxiv.org/abs/2602.18866