Timezone: »

Combating Label Noise in Deep Learning using Abstention
Sunil Thulasidasan · Tanmoy Bhattacharya · Jeff Bilmes · Gopinath Chennupati · Jamal Mohd-Yusof

Tue Jun 11 06:30 PM -- 09:00 PM (PDT) @ Pacific Ballroom #9

We introduce a novel method to combat label noise when training deep neural networks for classification. We propose a loss function that permits abstention during training thereby allowing the DNN to abstain on confusing samples while continuing to learn and improve classification performance on the non-abstained samples. We show how such a deep abstaining classifier (DAC) can be used for robust learning in the presence of different types of label noise. In the case of structured or systematic label noise – where noisy training labels or confusing examples are correlated with underlying features of the data– training with abstention enables representation learning for features that are associated with unreliable labels. In the case of unstructured (arbitrary) label noise, abstention during training enables the DAC to be used as an effective data cleaner by identifying samples that are likely to have label noise. We provide analytical results on the loss function behavior that enable dynamic adaption of abstention rates based on learning progress during training. We demonstrate the utility of the deep abstaining classifier for various image classification tasks under different types of label noise; in the case of arbitrary label noise, we show significant im- provements over previously published results on multiple image benchmarks.

Author Information

Sunil Thulasidasan (Los Alamos National Laboratory & University of Washington)
Tanmoy Bhattacharya (Los Alamos National Laboratory)
Jeff Bilmes (UW)
Gopinath Chennupati (Los Alamos National Laboratory)
Jamal Mohd-Yusof (Los Alamos National Laboratory)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors