[PDF] Learning Discriminative Dynamics with Label Corruption for Noisy Label Detection | Semantic Scholar (2024)

[PDF] Semantic Reader

A training-free solution to detect corrupted labels by checking the noisy label consensuses of nearby features and a ranking-based approach that scores each instance and filters out a guaranteed number of instances that are likely to be corrupted.

A general framework to train CNNs with only a limited number of clean labels and millions of easily obtained noisy labels is introduced and the relationships between images, class labels and label noises are model with a probabilistic graphical model and further integrate it into an end-to-end deep learning system.

PGDF (Prior Guided Denoising Framework), a novel framework to learn a deep model to suppress noise by generating the samples' prior knowledge, which is integrated into both dividing samples step and semi-supervised step, is proposed and demonstrates substantial improvements over state-of-the-art methods.

It is shown that the real-world noise patterns impose new and outstanding challenges as compared to synthetic label noise, and the availability of these two datasets would facilitate the development and evaluation of future learning with noisy label solutions.

A novel detector for filtering label noise that focuses on each data's latent representation dynamics and measure the alignment between the latent distribution and each representation using the eigendecomposition of the data gram matrix, providing a robust detector with derivative-free simple methods having theoretical guarantees.

These experiments establish that state-of-the-art convolutional networks for image classification trained with stochastic gradient methods easily fit a random labeling of the training data, and corroborate these experimental findings with a theoretical construction showing that simple depth two neural networks already have perfect finite sample expressivity.

This paper proposes CORES^2 (COnfidence REgularized Sample Sieve), which progressively sieves out corrupted samples and provides a generic machinery for anatomizing noisy datasets and a flexible interface for various robust training techniques to further improve the performance.

It is proved that early learning and memorization are fundamental phenomena in high-dimensional classification tasks, even in simple linear models, and a new technique for noisy classification tasks is developed, which exploits the progress of the early learning phase.

This work proposes DivideMix, a novel framework for learning with noisy labels by leveraging semi-supervised learning techniques, which models the per-sample loss distribution with a mixture model to dynamically divide the training data into a labeled set with clean samples and an unlabeled set with noisy samples.

A new method to identify overly ambiguous or outrightly mislabeled samples and mitigate their impact when training neural networks is introduced, at the heart of which is the Area Under the Margin (AUM) statistic.

...

...