Saved in:
Bibliographic Details
Main Authors: Chhabra, Sachin, Venkateswara, Hemanth, Li, Baoxin
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.05307
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909773774979072
author Chhabra, Sachin
Venkateswara, Hemanth
Li, Baoxin
author_facet Chhabra, Sachin
Venkateswara, Hemanth
Li, Baoxin
contents Training neural networks with one-hot target labels often results in overconfidence and overfitting. Label smoothing addresses this issue by perturbing the one-hot target labels by adding a uniform probability vector to create a regularized label. Although label smoothing improves the network's generalization ability, it assigns equal importance to all the non-target classes, which destroys the inter-class relationships. In this paper, we propose a novel label regularization training strategy called Label Smoothing++, which assigns non-zero probabilities to non-target classes and accounts for their inter-class relationships. Our approach uses a fixed label for the target class while enabling the network to learn the labels associated with non-target classes. Through extensive experiments on multiple datasets, we demonstrate how Label Smoothing++ mitigates overconfident predictions while promoting inter-class relationships and generalization capabilities.
format Preprint
id arxiv_https___arxiv_org_abs_2509_05307
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Label Smoothing++: Enhanced Label Regularization for Training Neural Networks
Chhabra, Sachin
Venkateswara, Hemanth
Li, Baoxin
Computer Vision and Pattern Recognition
Training neural networks with one-hot target labels often results in overconfidence and overfitting. Label smoothing addresses this issue by perturbing the one-hot target labels by adding a uniform probability vector to create a regularized label. Although label smoothing improves the network's generalization ability, it assigns equal importance to all the non-target classes, which destroys the inter-class relationships. In this paper, we propose a novel label regularization training strategy called Label Smoothing++, which assigns non-zero probabilities to non-target classes and accounts for their inter-class relationships. Our approach uses a fixed label for the target class while enabling the network to learn the labels associated with non-target classes. Through extensive experiments on multiple datasets, we demonstrate how Label Smoothing++ mitigates overconfident predictions while promoting inter-class relationships and generalization capabilities.
title Label Smoothing++: Enhanced Label Regularization for Training Neural Networks
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.05307