Saved in:
Bibliographic Details
Main Authors: Ahmad, Touqeer, Kalan, Mohammadreza M., Portier, François, Stupfler, Gilles
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.20472
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908607096815616
author Ahmad, Touqeer
Kalan, Mohammadreza M.
Portier, François
Stupfler, Gilles
author_facet Ahmad, Touqeer
Kalan, Mohammadreza M.
Portier, François
Stupfler, Gilles
contents Synthetic oversampling of minority examples using SMOTE and its variants is a leading strategy for addressing imbalanced classification problems. Despite the success of this approach in practice, its theoretical foundations remain underexplored. We develop a theoretical framework to analyze the behavior of SMOTE and related methods when classifiers are trained on synthetic data. We first derive a uniform concentration bound on the discrepancy between the empirical risk over synthetic minority samples and the population risk on the true minority distribution. We then provide a nonparametric excess risk guarantee for kernel-based classifiers trained using such synthetic data. These results lead to practical guidelines for better parameter tuning of both SMOTE and the downstream learning algorithm. Numerical experiments are provided to illustrate and support the theoretical findings
format Preprint
id arxiv_https___arxiv_org_abs_2510_20472
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Concentration and excess risk bounds for imbalanced classification with synthetic oversampling
Ahmad, Touqeer
Kalan, Mohammadreza M.
Portier, François
Stupfler, Gilles
Machine Learning
Synthetic oversampling of minority examples using SMOTE and its variants is a leading strategy for addressing imbalanced classification problems. Despite the success of this approach in practice, its theoretical foundations remain underexplored. We develop a theoretical framework to analyze the behavior of SMOTE and related methods when classifiers are trained on synthetic data. We first derive a uniform concentration bound on the discrepancy between the empirical risk over synthetic minority samples and the population risk on the true minority distribution. We then provide a nonparametric excess risk guarantee for kernel-based classifiers trained using such synthetic data. These results lead to practical guidelines for better parameter tuning of both SMOTE and the downstream learning algorithm. Numerical experiments are provided to illustrate and support the theoretical findings
title Concentration and excess risk bounds for imbalanced classification with synthetic oversampling
topic Machine Learning
url https://arxiv.org/abs/2510.20472