Intrinsic Dimensionality as a Model-Free Measure of Class Imbalance

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Eser, Çağrı, Baltacı, Zeynep Sonat, Akbaş, Emre, Kalkan, Sinan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912836320493568
author Eser, Çağrı
Baltacı, Zeynep Sonat
Akbaş, Emre
Kalkan, Sinan
author_facet Eser, Çağrı
Baltacı, Zeynep Sonat
Akbaş, Emre
Kalkan, Sinan
contents Imbalance in classification tasks is commonly quantified by the cardinalities of examples across classes. This, however, disregards the presence of redundant examples and inherent differences in the learning difficulties of classes. Alternatively, one can use complex measures such as training loss and uncertainty, which, however, depend on training a machine learning model. Our paper proposes using data Intrinsic Dimensionality (ID) as an easy-to-compute, model-free measure of imbalance that can be seamlessly incorporated into various imbalance mitigation methods. Our results across five different datasets with a diverse range of imbalance ratios show that ID consistently outperforms cardinality-based re-weighting and re-sampling techniques used in the literature. Moreover, we show that combining ID with cardinality can further improve performance. Our code and models are available at https://github.com/cagries/IDIM.
format Preprint
id arxiv_https___arxiv_org_abs_2511_10475
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Intrinsic Dimensionality as a Model-Free Measure of Class Imbalance
Eser, Çağrı
Baltacı, Zeynep Sonat
Akbaş, Emre
Kalkan, Sinan
Machine Learning
Computer Vision and Pattern Recognition
Imbalance in classification tasks is commonly quantified by the cardinalities of examples across classes. This, however, disregards the presence of redundant examples and inherent differences in the learning difficulties of classes. Alternatively, one can use complex measures such as training loss and uncertainty, which, however, depend on training a machine learning model. Our paper proposes using data Intrinsic Dimensionality (ID) as an easy-to-compute, model-free measure of imbalance that can be seamlessly incorporated into various imbalance mitigation methods. Our results across five different datasets with a diverse range of imbalance ratios show that ID consistently outperforms cardinality-based re-weighting and re-sampling techniques used in the literature. Moreover, we show that combining ID with cardinality can further improve performance. Our code and models are available at https://github.com/cagries/IDIM.
title Intrinsic Dimensionality as a Model-Free Measure of Class Imbalance
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.10475