Deep Self-Learning From Noisy Labels

By Jiangfan Han, Ping Luo, Xiaogang Wang. 2019

321 citations recorded by the source corpus.

ConvNets achieve good results when training from clean data, but learning from noisy labels significantly degrades performances and remains challenging. Unlike previous works constrained by many conditions, making them infeasible to real noisy cases, this work presents a novel deep self-learning framework to train a robust network on the real noisy datasets without extra supervision. The proposed approach has several appealing benefits. (1) Different from most existing work, it does not rely on any assumption on the distribution of the noisy labels, making it robust to real noises. (2) It does not need extra clean supervision or accessorial network to help training. (3) A self-learning framework is proposed to train the network in an iterative end-to-end manner, which is effective and efficient. Extensive experiments in challenging benchmarks such as Clothing1M and Food101-N show that our approach outperforms its counterparts in all empirical settings.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Continual Learning on Noisy Data Streams via Self-Purified Replay (2021, background): “Techniques that repair labels [39, 80, 50, 74, 27, 56] or directly learn them [76, 88] are also viable options for learning from noisy labeled data.”
DualGraph: A graph-based method for reasoning about label noise (2021, methodology): “For experiments on Clothing1M, we adopt the following three settings by following previous work [6].”
Identifying Mislabeled Data using the Area Under the Margin Ranking (2020, methodology): “) We note that our method and many baselines can be used in conjunction with semi-supervised learning [33, 44], pseudo-labeling [20, 66], or MixUp [3, 64] to improve noisy-training performance.”
Improving Generalization by Controlling Label-Noise Information in Neural Network Weights (2020, background): “Others seek to detect incorrect examples and relabel them (Reed et al., 2014; Tanaka et al., 2018; Ma et al., 2018; Han et al., 2019; Arazo et al., 2019).”
Confident Learning: Estimating Uncertainty in Dataset Labels (2019, background): “…loss (Patrini et al., 2016, 2017b; Sukhbaatar et al., 2014), imputation (Li et al., 2017; Amjad et al., 2017), crowd-based approaches (Zhang et al., 2017b; Dawid and Skene, 1979; Ratner et al., 2016), error removal (Wang et al., 2018; Northcutt et al., 2017), or fixing labels (Han et al., 2019).”
Deep learning with noisy labels: exploring techniques and remedies in medical image analysis (2019, unknown): “[113] improved upon CleanNet in several ways.”
Robust Multi-View Clustering With Incomplete Information (2022, methodology): “Based on the way of achieving robustness, the existing works could be roughly grouped into four categories, namely, i) robust loss based methods [30], [32], which design a loss function which is tolerant to noisy labels; ii) robust architecture based methods [33], [34], which modify the network arch”
Prototypical Pseudo Label Denoising and Target Structure Learning for Domain Adaptive Semantic Segmentation (2021, methodology): “Our proposed pseudo label denoising is more close to [23] which online corrects the incorrect labels according to the prototypes determined by some complex heuristics.”
Delving Deep Into Label Smoothing (2020, unknown): “Han et al. [45] designed the label correction phase and performed the training phase and label correction phase iteratively.”
Mutual Mean-Teaching: Pseudo Label Refinery for Unsupervised Domain Adaptation on Person Re-identification (2020, background): “(Veit et al., 2017; Lee et al., 2018; Li et al., 2017; Han et al., 2019) attempted to correct the noisy labels directly, while the clean set required by such methods limits their generalization on real-world applications.”
Image Classification with Deep Learning in the Presence of Noisy Labels: A Survey (2019, methodology): “[63] constructs prototypes that are able to represent deep feature distribution of the corresponding class, for each class.”
Combating Noisy Labels by Agreement: A Joint Training Method with Co-Regularization (2020, background): “In addition to the afore-mentioned approaches, there are some other deep learning solutions [13, 17] to deal with noisy labels, including pseudo-label based [35, 40] and robust loss based approaches [28, 46].”