Awesome-Noisy-Labels

Noisy label handling survey

This repository provides a comprehensive survey and collection of papers on techniques for handling noisy labels in machine learning with deep neural networks.

A Survey

GitHub
545 stars
5 watching
31 forks
last commit: almost 2 years ago
Linked from 1 awesome list


Learning from Noisy Labels with Deep Neural Networks: A Survey
arxiv version [ ]

Learning from Noisy Labels with Deep Neural Networks: A Survey / A. Robust Architecture
Webly supervised learning of convolutional networks
Official (Caffe) 7 almost 7 years ago
Training convolutional networks with noisy labels
Unofficial (Keras) 6 almost 8 years ago
Learning deep networks from noisy labels with dropout regularization
Official (MATLAB) 11 over 6 years ago
Training deep neural-networks based on unreliable labels
Unofficial (Chainer) 5 almost 8 years ago
Training deep neural-networks using a noise adaptation layer
Official (Keras) 118 over 7 years ago
Learning from massive noisy labeled data for image classification
Official (Caffe) 116 almost 6 years ago
Masking: A new perspective of noisy supervision
Official (TensorFlow) 54 about 6 years ago
Deep learning from noisy image labels with quality embedding
Robust inference via generative classifiers for handling noisy labels
Official (PyTorch) 33 over 5 years ago

Learning from Noisy Labels with Deep Neural Networks: A Survey / B. Robust Regularization
Deep bilevel learning
Official (TensorFlow) 11 almost 5 years ago
Learning from noisy labels by regularized estimation of annotator confusion
Official (TensorFlow)
Using pre-training can improve model robustness and uncertainty
Official (PyTorch)
Can gradient clipping mitigate label noise?
Unofficial (PyTorch) 14 5 months ago
Wasserstein adversarial regularization (WAR) on label noise
Robust early-learning: Hindering the memorization of noisy labels
Official (PyTorch) 75 over 3 years ago
When Optimizing f-Divergence is Robust with Label Noise
Official (PyTorch)
Learning with Noisy Labels via Sparse Regularization
Official (PyTorch) 46 almost 3 years ago
Open-set Label Noise Can Improve Robustness Against Inherent Label Noise
Official (PyTorch) 19 over 2 years ago
Explaining and harnessing adversarial examples
Unofficial (PyTorch)
Regularizing neural networks by penalizing confident output distributions
Unofficial (PyTorch) 2,196 3 months ago
Mixup: Beyond empirical risk minimization
Official (PyTorch) 1,170 over 3 years ago
Augmentation Strategies for Learning with Noisy Labels
Official (PyTorch) 113 about 3 years ago
AutoDO: Robust AutoAugment for Biased Data With Label Noise via Scalable Probabilistic Implicit Differentiation
Official (PyTorch) 24 about 2 years ago

Learning from Noisy Labels with Deep Neural Networks: A Survey / C. Robust Loss Function
Robust loss functions under label noise for deep neural networks
Symmetric cross entropy for robust learning with noisy labels
Official (Keras)
Generalized cross entropy loss for training deep neural networks with noisy labels
Unofficial (PyTorch) 126 about 5 years ago
Curriculum loss: Robust learning and generalization against label corruption
Normalized loss functions for deep learning with noisy labels
Official (PyTorch) 134 7 months ago
Peer loss functions: Learning from noisy labels without knowing noise rates
Official (PyTorch) 35 almost 5 years ago
Learning Cross-Modal Retrieval with Noisy Labels
Official (Pytorch) 13 almost 4 years ago
A Second-Order Approach to Learning With Instance-Dependent Label Noise
Official (PyTorch) 47 about 2 years ago
An Information Fusion Approach to Learning with Instance-Dependent Label Noise

Learning from Noisy Labels with Deep Neural Networks: A Survey / D. Loss Adjustment
Making deep neural networks robust to label noise: A loss correction approach
Official (Keras) 90 over 4 years ago
Using trusted data to train deep networks on labels corrupted by severe noise
Official (PyTorch) 86 about 6 years ago
Are anchor points really indispensable in label-noise learning?
Official (PyTorch) 97 over 3 years ago
Dual T: Reducing estimation error for transition matrix in label-noise learning
Tackling Instance-Dependent Label Noise via a Universal Probabilistic Model
Official (PyTorch) 9 over 3 years ago
Multiclass learning with partially corrupted labels
Unofficial (PyTorch) 39 over 5 years ago
Active Bias: Training more accurate neural networks by emphasizing high variance samples
Unofficial (TensorFlow) 20 almost 6 years ago
Training deep neural networks on noisy labels with bootstrapping
Unofficial (Keras) 22 almost 4 years ago
Dimensionality-driven learning with noisy labels
Official (Keras) 58 7 months ago
Unsupervised label noise modeling and loss correction
Official (PyTorch) 221 over 4 years ago
Self-adaptive training: beyond empirical risk minimization
Official (PyTorch) 127 over 3 years ago
Error-bounded correction of noisy labels
Official (PyTorch) 22 about 2 years ago
Beyond class-conditional assumption: A primary attempt to combat instancedependent label noise
Official (PyTorch) 35 over 3 years ago
Learning to learn from weak supervision by full supervision
Unofficial (TensorFlow) 4 about 7 years ago
Learning from noisy labels with distillation
Learning to reweight examples for robust deep learning
Official (TensorFlow) 269 almost 6 years ago
Meta-Weight-Net: Learning an explicit mapping for sample weighting
Official (PyTorch) 284 about 3 years ago
Distilling effective supervision from severe label noise
Official (TensorFlow) 34,478 about 1 month ago
Meta label correction for noisy label learning
Official (PyTorch)
Adaptive Label Noise Cleaning with Meta-Supervision for Deep Face Recognition

Learning from Noisy Labels with Deep Neural Networks: A Survey / E. Sample Selection
Decoupling when to update from how to update
Official (TensorFlow)
MentorNet: Learning data-driven curriculum for very deep neural networks on corrupted labels
Official (TensorFlow) 322 almost 2 years ago
Co-teaching: Robust training of deep neural networks with extremely noisy labels
Official (PyTorch) 494 over 3 years ago
How does disagreement help generalization against label corruption?
Official (PyTorch) 21 over 5 years ago
Jo-SRC: A Contrastive Approach for Combating Noisy Labels
Official (PyTorch) 5 almost 4 years ago
Iterative learning with open-set noisy labels
Official (Keras)
Learning with bad training data via iterative trimmed loss minimization
Official (GluonCV) 7 over 5 years ago
Understanding and utilizing deep neural networks trained with noisy labels
Official (Keras) 90 about 4 years ago
O2U-Net: A simple noisy label detection approach for deep neural networks
Unofficial (PyTorch) 77 over 2 years ago
How does early stopping can help generalization against label noise?
Official (Tensorflow)
A topological filter for learning with label noise
Official (PyTorch) 30 about 2 years ago
Learning with Instance-Dependent Label Noise: A Sample Sieve Approach
Official (PyTorch) 36 almost 4 years ago
FINE Samples for Learning with Noisy Labels
Official (PyTorch) 39 about 3 years ago
Sample Selection with Uncertainty of Losses for Learning with Noisy Labels
SELFIE: Refurbishing unclean samples for robust deep learning
Official (TensorFlow) 50 about 5 years ago
SELF: Learning to filter noisy labels with self-ensembling
DivideMix: Learning with noisy labels as semi-supervised learning
Official (PyTorch) 546 over 4 years ago
Robust curriculum learning: from clean label detection to noisy label self-correction
Understanding and Improving Early Stopping for Learning with Noisy Labels
Official (PyTorch) 29 about 2 years ago

Learning from Noisy Labels with Deep Neural Networks: A Survey / Learning from Noisy Labels towards Realistic Scenarios
Online Continual Learning on a Contaminated Data Stream with Blurry Task Boundaries , CVPR 2022, [ ] This paper addresses the problem of noisy labels in the online continual learning setup
Learning from Multiple Annotator Noisy Labels via Sample-wise Label Fusion , ECCV 2022, [ ] This paper addresses the scenario in which each data instance has multiple noisy labels from annotators (instead of a single label)

Backlinks from these awesome lists:

0