Learning from noisy labels with deep neural networks: A survey

H Song, M Kim, D Park, Y Shin… - IEEE transactions on …, 2022‏ - ieeexplore.ieee.org
Deep learning has achieved remarkable success in numerous domains with help from large
amounts of big data. However, the quality of data labels is a concern because of the lack of …

The METRIC-framework for assessing data quality for trustworthy AI in medicine: a systematic review

D Schwabe, K Becker, M Seyferth, A Klaß… - NPJ Digital …, 2024‏ - nature.com
The adoption of machine learning (ML) and, more specifically, deep learning (DL)
applications into all major areas of our lives is underway. The development of trustworthy AI …

Robust training under label noise by over-parameterization

S Liu, Z Zhu, Q Qu, C You - International Conference on …, 2022‏ - proceedings.mlr.press
Recently, over-parameterized deep networks, with increasingly more network parameters
than training samples, have dominated the performances of modern machine learning …

Benchmarking uncertainty disentanglement: Specialized uncertainties for specialized tasks

B Mucsányi, M Kirchhof, SJ Oh - Advances in Neural …, 2025‏ - proceedings.neurips.cc
Uncertainty quantification, once a singular task, has evolved into a spectrum of tasks,
including abstained prediction, out-of-distribution detection, and aleatoric uncertainty …

Estimating noise transition matrix with label correlations for noisy multi-label learning

S Li, X **a, H Zhang, Y Zhan… - Advances in Neural …, 2022‏ - proceedings.neurips.cc
In label-noise learning, the noise transition matrix, bridging the class posterior for noisy and
clean data, has been widely exploited to learn statistically consistent classifiers. The …

Label-free node classification on graphs with large language models (llms)

Z Chen, H Mao, H Wen, H Han, W **, H Zhang… - arxiv preprint arxiv …, 2023‏ - arxiv.org
In recent years, there have been remarkable advancements in node classification achieved
by Graph Neural Networks (GNNs). However, they necessitate abundant high-quality labels …

Fine-grained classification with noisy labels

Q Wei, L Feng, H Sun, R Wang… - Proceedings of the …, 2023‏ - openaccess.thecvf.com
Learning with noisy labels (LNL) aims to ensure model generalization given a label-
corrupted training set. In this work, we investigate a rarely studied scenario of LNL on fine …

Targeted representation alignment for open-world semi-supervised learning

R **ao, L Feng, K Tang, J Zhao, Y Li… - Proceedings of the …, 2024‏ - openaccess.thecvf.com
Abstract Open-world Semi-Supervised Learning aims to classify unlabeled samples utilizing
information from labeled data while unlabeled samples are not only from the labeled known …

Asymmetric loss functions for noise-tolerant learning: Theory and applications

X Zhou, X Liu, D Zhai, J Jiang… - IEEE Transactions on …, 2023‏ - ieeexplore.ieee.org
Supervised deep learning has achieved tremendous success in many computer vision
tasks, which however is prone to overfit noisy labels. To mitigate the undesirable influence of …

Robust data pruning under label noise via maximizing re-labeling accuracy

D Park, S Choi, D Kim, H Song… - Advances in Neural …, 2023‏ - proceedings.neurips.cc
Data pruning, which aims to downsize a large training set into a small informative subset, is
crucial for reducing the enormous computational costs of modern deep learning. Though …