• Home
  • Search
  • Stochastic Feature Averaging for Learning with Long-Tailed Noisy Labels
  • Cite Icon8
  • https://doi.org/10.24963/ijcai.2023/434Copy DOI Icon

Stochastic Feature Averaging for Learning with Long-Tailed Noisy Labels

  • Aug 1, 2023
  • Hao-Tian Li +7 more
Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Deep neural networks have shown promising results on a wide variety of tasks using large-scale and well-annotated training datasets. However, data collected from real-world applications can suffer from two prevalent biases, i.e., long-tailed class distribution and label noise. Previous efforts on long-tailed learning and label-noise learning can only address a single type of data bias, leading to a severe deterioration of their performance. In this paper, we propose a distance-based sample selection algorithm called Stochastic Feature Averaging (SFA), which fits a Gaussian using the exponential running average of class centroids to capture uncertainty in representation space due to label noise and data scarcity. With SFA, we detect noisy samples based on their distances to class centroids sampled from this Gaussian distribution. Based on the identified clean samples, we then propose to train an auxiliary balanced classifier to improve the generalization for the minority class and facilitate the update of Gaussian parameters. Extensive experimental results show that SFA can enhance the performance of existing methods on both simulated and real-world datasets. Further, we propose to combine SFA with the sample-selection approach, distribution-robust, and noise-robust loss functions, resulting in significant improvement in performance over the baselines. Our code is available at https://github.com/HotanLee/SFA

Similar Papers
  • Research Article
  • Citations6

Suppressing label noise in medical image classification using mixup attention and self-supervised learning

  • May 08, 2024
  • Physics in Medicine & Biology
  • Mengdi Gao +5
  • Book Chapter
  • Citations4

How Does Heterogeneous Label Noise Impact Generalization in Neural Nets?

  • Jan 01, 2021
  • Bidur Khanal +1
  • Research Article
  • Citations11

Gradient-aware learning for joint biases: Label noise and class imbalance

  • Dec 18, 2023
  • Neural Networks
  • Shichuan Zhang +4
  • Research Article
  • Citations358

Image classification with deep learning in the presence of noisy labels: A survey

  • Jan 13, 2021
  • Knowledge-Based Systems
  • Görkem Algan +1
  • Research Article
  • Citations33

A Survey on Classifying Big Data with Label Noise

  • Nov 23, 2022
  • Journal of Data and Information Quality
  • Justin M Johnson +1
  • Research Article
  • Citations22

A membership-based resampling and cleaning algorithm for multi-class imbalanced overlapping data

  • Nov 11, 2023
  • Expert Systems With Applications
  • Tingting Ma +2
  • Research Article
  • Citations11

A Convergence Path to Deep Learning on Noisy Labels.

  • Apr 01, 2024
  • IEEE Transactions on Neural Networks and Learning Systems
  • Defu Liu +2
  • Conference Article
  • Citations7

Graph-Based Self-Learning for Robust Person Re-identification

  • Jan 01, 2023
  • Yuqiao Xian +4
  • Research Article
  • Citations21

Dynamics-aware loss for learning with label noise

  • Aug 01, 2023
  • Pattern Recognition
  • Xiu-Chuan Li +5
  • Research Article
  • Citations21

Meta-Learning Guided Label Noise Distillation for Robust Signal Modulation Classification

  • Jan 01, 2024
  • IEEE Internet of Things Journal
  • Xiaoyang Hao +3
  • Conference Article
  • Citations5

Training Classifiers that are Universally Robust to All Label Noise Levels

  • Jul 18, 2021
  • Jingyi Xu +2
  • Research Article
  • Citations110

Improving Medical Images Classification With Label Noise Using Dual-Uncertainty Estimation.

  • Jun 01, 2022
  • IEEE Transactions on Medical Imaging
  • Lie Ju +7
  • PDF
  • Research Article
  • Citations209

Effect of Training Class Label Noise on Classification Performances for Land Cover Mapping with Satellite Image Time Series

  • Feb 18, 2017
  • Remote Sensing
  • Charlotte Pelletier +5
  • Conference Article
  • Citations23

Noisy Concurrent Training for Efficient Learning under Label Noise

  • Jan 01, 2021
  • Fahad Sarfraz +2
  • Book Chapter
  • Citations4

Self-paced Robust Deep Face Recognition with Label Noise

  • Jan 01, 2019
  • Pengfei Zhu +2
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.