Corpus ID: 6546734

Robust Loss Functions under Label Noise for Deep Neural Networks

@article{Ghosh2017RobustLF,
  title={Robust Loss Functions under Label Noise for Deep Neural Networks},
  author={Aritra Ghosh and H. Kumar and P. S. Sastry},
  journal={ArXiv},
  year={2017},
  volume={abs/1712.09482}
}
  • Aritra Ghosh, H. Kumar, P. S. Sastry
  • Published 2017
  • Computer Science, Mathematics
  • ArXiv
  • In many applications of classifier learning, training data suffers from label noise. [...] Key Result Through experiments, we illustrate the robustness of risk minimization with such loss functions for learning neural networks.Expand Abstract
    Robust Loss Functions for Learning Multi-class Classifiers
    2
    Normalized Loss Functions for Deep Learning with Noisy Labels
    2
    Image Classification with Deep Learning in the Presence of Noisy Labels: A Survey
    9
    On Robustness of Neural Architecture Search Under Label Noise
    Can Cross Entropy Loss Be Robust to Label Noise?
    1
    Theoretical Guarantees of Deep Embedding Losses Under Label Noise
    1
    Learning from Noisy Labels with Deep Neural Networks: A Survey
    2
    Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labels
    241

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 33 REFERENCES
    Making risk minimization tolerant to label noise
    95
    Learning with Noisy Labels
    508
    Noise Tolerance Under Risk Minimization
    125