Corpus ID: 53217518

FUNN: Flexible Unsupervised Neural Network

@article{Vigouroux2018FUNNFU,
  title={FUNN: Flexible Unsupervised Neural Network},
  author={David Vigouroux and S. Picard},
  journal={ArXiv},
  year={2018},
  volume={abs/1811.01749}
}
  • David Vigouroux, S. Picard
  • Published 2018
  • Mathematics, Computer Science
  • ArXiv
  • Deep neural networks have demonstrated high accuracy in image classification tasks. However, they were shown to be weak against adversarial examples: a small perturbation in the image which changes the classification output dramatically. In recent years, several defenses have been proposed to solve this issue in supervised classification tasks. We propose a method to obtain robust features in unsupervised learning tasks against adversarial attacks. Our method differs from existing solutions by… CONTINUE READING

    References

    SHOWING 1-10 OF 24 REFERENCES
    Learning to Attack: Adversarial Transformation Networks
    • 61
    • PDF
    Simple Black-Box Adversarial Perturbations for Deep Networks
    • 125
    • Highly Influential
    • PDF
    On Detecting Adversarial Perturbations
    • 473
    • PDF
    Adversarial Defense based on Structure-to-Signal Autoencoders
    • 7
    • PDF
    MagNet: A Two-Pronged Defense against Adversarial Examples
    • 509
    • Highly Influential
    • PDF
    Generative Adversarial Trainer: Defense to Adversarial Perturbations with GAN
    • 80
    • PDF
    DeepFool: A Simple and Accurate Method to Fool Deep Neural Networks
    • 1,829
    • PDF
    Distillation as a Defense to Adversarial Perturbations Against Deep Neural Networks
    • 1,437
    • PDF