Learning from massive noisy labeled data for image classification

By Tong Xiao, Tian Xia, Yi Yang, Chang Huang, Xiaogang Wang. 2015

1,303 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Industrial and Manufacturing Engineering, Evaluation & Benchmarks

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

In Industrial and Manufacturing Engineering

In Evaluation & Benchmarks

Citations in the corpus

Learning Visual N-Grams from Web Data (2016, methodology): “Various other studies have used data from photo-sharing websites such as Flickr for training; for instance, to train hierarchical topic models [38] or multiple-instance learning SVMs [39], to learn label distribution models [12, 64], to finetune pretrained convolutional networks [24], and to train w”
On Robustness of Neural Architecture Search Under Label Noise (2020, methodology): “Another group of approaches treats the true but unknown labels as latent variables and the noisy labels as observed variables so that EMlike algorithms can be used to learn the true label distribution of the dataset (Xiao et al., 2015; Vahdat, 2017; Khetan et al., 2018).”
A Survey on Data Collection for Machine Learning: A Big Data - AI Integration Perspective (2018, background): “Improve Model Robust Against Noise [167], [168], [169], [170], [171] Transfer Learning [172], [173], [174], [175], [176], [177], [178] 1340 IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, VOL.”
The Unreasonable Effectiveness of Noisy Data for Fine-Grained Recognition (2015, methodology): “In this section we provide an analysis of the imagery publicly available for finegrained recognition, which we collect via web search.1 We describe its quantity, distribution, and levels of noise, reporting each on multiple fine-grained domains.”
Learning from Noisy Large-Scale Datasets with Minimal Supervision (2017, neutral): “[32] that proposes an image-conditioned”
Learning to Reweight Examples for Robust Deep Learning (2018, methodology): “…community (Natarajan et al., 2013; Angluin & Laird, 1988) and practical methods have also been proposed (Reed et al., 2014; Sukhbaatar & Fergus, 2014; Xiao et al., 2015; Azadi et al., 2016; Goldberger & Ben-Reuven, 2017; Li et al., 2017; Jiang et al., 2017; Vahdat, 2017; Hendrycks et al., 2018).”
Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labels (2018, background): “Other approaches include treating the true labels as a latent variable and the noisy labels as an observed variable so that EM-like algorithms can be used to learn true label distribution of the dataset [40, 17, 36].”
Noise Estimation Using Density Estimation for Self-Supervised Multimodal Learning (2020, methodology): “…(a) formulating explicit or implicit noise models to characterize the distribution of noisy and true labels using neural networks (Goldberger and BenReuven 2017; Jiang et al. 2018; Sukhbaatar et al. 2015), graphical models (Xiao et al. 2015; Li et al. 2017), etc. and (b) using correction methods.”
Using Trusted Data to Train Deep Networks on Labels Corrupted by Severe Noise (2018, methodology): “This assumption has been leveraged by others for the purpose of label noise robustness, most notably [24, 9, 25, 19].”
Robot Learning in Homes: Improving Generalization and Reducing Dataset Bias (2018, background): “This means that creating models that can account for and correct noise [18, 38] are valuable.”
Unsupervised label noise modeling and loss correction (2019, methodology): “Other approaches seek to relabel the noisy samples by modeling their noise through directed graphical models (Xiao et al., 2015), Conditional Random Fields (Vahdat, 2017), or CNNs (Veit et al.”
CleanNet: Transfer Learning for Scalable Image Classifier Training with Label Noise (2017, methodology): “Comparing to classiﬁcation ﬁltering at 16.60% on Food101-N and 23.55% on Clothing1M, the results demonstrate effectiveness of adding veriﬁcation labels for human supervision for label noise detection.”
Robust Loss Functions for Learning Multi-class Classifiers (2018, neutral): “A more general model similar to this is in [14].”
Learning with Noisy Labels Revisited: A Study Using Real-World Human Annotations (2021, background): “, 2014], Clothing-1M [Xiao et al., 2015], WebVision [Li et al.”
Synthetic vs Real: Deep Learning on Controlled Noise (2019, background): “…the most common type consists of images that are automatically tagged according to their surrounding texts either by directly crawling the web pages, e.g. Clothing-1M (Xiao et al., 2015), Instagram (Mahajan et al., 2018), or by querying an image search engine, e.g. WebVision (Li et al., 2017a).”