An Efficient and Provable Approach for Mixture Proportion Estimation Using Linear Independence Assumption

Yu, Xiyu; Liu, Tongliang; Gong, Maoguo; Batmanghelich, Kayhan; Tao, Dacheng

doi:10.1109/cvpr.2018.00471

Cited by 32 publications

(18 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…On the other hand, although it is costly to annotate a very large-scale dataset, a small set of easily distinguishable observations are assumed to be available in practice. This assumption is also widely used in estimating transition probabilities in label noise problem [31] and class priors in semisupervised learning [37]. Therefore, in order to estimate Q, we manually assign true labels to 5 or 10 observations in each class.…”

Section: Estimating Qmentioning

confidence: 99%

Learning with Biased Complementary Labels

Liu

Gong

et al. 2018

Lecture Notes in Computer Science

Self Cite

133

165

View full text Add to dashboard Cite

In this paper, we study the classi cation problem in which we have access to easily obtainable surrogate for true labels, namely complementary labels, which specify classes that observations do not belong to. Let Y andȲ be the true and complementary labels, respectively. We rst model the annotation of complementary labels via transition probabilities P (Ȳ = i|Y = j), i = j ∈ {1, · · · , c}, where c is the number of classes. Previous methods implicitly assume that P (Ȳ = i|Y = j), ∀i = j, are identical, which is not true in practice because humans are biased toward their own experience. For example, as shown in Figure 1, if an annotator is more familiar with monkeys than prairie dogs when providing complementary labels for meerkats, she is more likely to employ "monkey" as a complementary label. We therefore reason that the transition probabilities will be di erent. In this paper, we propose a framework that contributes three main innovations to learning with biased complementary labels: (1) It estimates transition probabilities with no bias. (2) It provides a general method to modify traditional loss functions and extends standard deep neural network classi ers to learn with biased complementary labels. (3) It theoretically ensures that the classi er learned with complementary labels converges to the optimal one learned with true labels. Comprehensive experiments on several benchmark datasets validate the superiority of our method to current state-of-the-art methods.

show abstract

Section: Estimating Qmentioning

confidence: 99%

Learning with Biased Complementary Labels

Liu

Gong

et al. 2018

Lecture Notes in Computer Science

Self Cite

133

165

View full text Add to dashboard Cite

show abstract

“…Here, we assume the noise level τ is known and set R(T ) = 1 − min{ T T k τ, τ } with T k =10. If τ is not known in advanced, it can be inferred using validation sets [33,75]. As for performance measurement, we use test accuracy, i.e., test accuracy = (# of correct prediction) / (# of testing).…”

Section: Experiments On Synthetic Balanced Noisy Datasetsmentioning

confidence: 99%

Sample Selection with Uncertainty of Losses for Learning with Noisy Labels

Xia,

Liu,

Han

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

In learning with noisy labels, the sample selection approach is very popular, which regards small-loss data as correctly labeled during training. However, losses are generated on-the-y based on the model being trained with noisy labels, and thus large-loss data are likely but not certainly to be incorrect. There are actually two possibilities of a large-loss data point: (a) it is mislabeled, and then its loss decreases slower than other data, since deep neural networks "learn patterns rst"; (b) it belongs to an underrepresented group of data and has not been selected yet.In this paper, we incorporate the uncertainty of losses by adopting interval estimation instead of point estimation of losses, where lower bounds of the con dence intervals of losses derived from distribution-free concentration inequalities, but not losses themselves, are used for sample selection. In this way, we also give large-loss but less selected data a try; then, we can better distinguish between the cases (a) and (b) by seeing if the losses e ectively decrease with the uncertainty after the try. As a result, we can better explore underrepresented data that are correctly labeled but seem to be mislabeled at rst glance. Experiments demonstrate that the proposed method is superior to baselines and robust to a broad range of label noise types.

show abstract

“…where τ is the estimated noise rate, which can be inferred using validation sets [12,25]. The value of λ t decreases quickly at the first T epochs until reaching 1 − τ .…”

Section: Considering a K-class Classification Problem With The Noisy Training Datamentioning

confidence: 99%

Small-Vote Sample Selection for Label-Noise Learning

Yan

Xue

et al. 2021

Machine Learning and Knowledge Discovery in Databases. Research Track

View full text Add to dashboard Cite

The small-loss criterion is widely used in recent label-noise learning methods. However, such a criterion only considers the loss of each training sample in a mini-batch but ignores the loss distribution in the whole training set. Moreover, the selection of clean samples depends on a heuristic clean data rate. As a result, some noisy-labeled samples are easily identified as clean ones, and vice versa. In this paper, we propose a novel yet simple sample selection method, which mainly consists of a Hierarchical Voting Scheme (HVS) and an Adaptive Clean data rate Estimation Strategy (ACES), to accurately identify clean samples and noisy-labeled samples for robust learning. Specifically, we propose HVS to effectively combine the global vote and the local vote, so that both epoch-level and batch-level information is exploited to assign a hierarchical vote for each mini-batch sample. Based on HVS, we further develop ACES to adaptively estimate the clean data rate by leveraging a 1D Gaussian Mixture Model (GMM). Experimental results show that our proposed method consistently outperforms several state-of-theart label-noise learning methods on both synthetic and real-world noisy benchmark datasets.

show abstract

An Efficient and Provable Approach for Mixture Proportion Estimation Using Linear Independence Assumption

Cited by 32 publications

References 18 publications

Learning with Biased Complementary Labels

Learning with Biased Complementary Labels

Sample Selection with Uncertainty of Losses for Learning with Noisy Labels

Small-Vote Sample Selection for Label-Noise Learning

Contact Info

Product

Resources

About