Step-E: A Differentiable Data Cleaning Framework for Robust Learning with Noisy Labels

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Du, Wenzhang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911278404993024
author Du, Wenzhang
author_facet Du, Wenzhang
contents Training data collected in the wild often contain noisy labels and outliers that substantially degrade the performance and reliability of deep neural networks. While data cleaning is commonly applied as a separate preprocessing stage, such two-stage pipelines neither fully exploit feedback from the downstream model nor adapt to unknown noise patterns. We propose Step-E, a simple framework that integrates sample selection and model learning into a single optimization process. At each epoch, Step-E ranks samples by loss and gradually increases the fraction of high-loss examples that are excluded from gradient updates after a brief warm-up stage, yielding an online curriculum that focuses on easy and consistent examples and eventually ignores persistent outliers. On CIFAR-100N, Step-E improves the test accuracy of a ResNet-18 model from 43.3% (+/- 0.7%) to 50.4% (+/- 0.9%), clearly outperforming loss truncation, self-paced learning, and one-shot filtering while approaching the clean-label oracle at 60.5% (+/- 0.2%). On CIFAR-10N (aggre), Step-E also improves over the noisy baseline (85.3% vs. 83.9%) and nearly matches the clean-label oracle (85.9%), with only moderate training-time overhead.
format Preprint
id arxiv_https___arxiv_org_abs_2511_17040
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Step-E: A Differentiable Data Cleaning Framework for Robust Learning with Noisy Labels
Du, Wenzhang
Machine Learning
68T05
I.2.6; I.5.1
Training data collected in the wild often contain noisy labels and outliers that substantially degrade the performance and reliability of deep neural networks. While data cleaning is commonly applied as a separate preprocessing stage, such two-stage pipelines neither fully exploit feedback from the downstream model nor adapt to unknown noise patterns. We propose Step-E, a simple framework that integrates sample selection and model learning into a single optimization process. At each epoch, Step-E ranks samples by loss and gradually increases the fraction of high-loss examples that are excluded from gradient updates after a brief warm-up stage, yielding an online curriculum that focuses on easy and consistent examples and eventually ignores persistent outliers. On CIFAR-100N, Step-E improves the test accuracy of a ResNet-18 model from 43.3% (+/- 0.7%) to 50.4% (+/- 0.9%), clearly outperforming loss truncation, self-paced learning, and one-shot filtering while approaching the clean-label oracle at 60.5% (+/- 0.2%). On CIFAR-10N (aggre), Step-E also improves over the noisy baseline (85.3% vs. 83.9%) and nearly matches the clean-label oracle (85.9%), with only moderate training-time overhead.
title Step-E: A Differentiable Data Cleaning Framework for Robust Learning with Noisy Labels
topic Machine Learning
68T05
I.2.6; I.5.1
url https://arxiv.org/abs/2511.17040