Corpus ID: 195345734

Computer Vision with a Single (Robust) Classifier

@article{Santurkar2019ComputerVW,
  title={Computer Vision with a Single (Robust) Classifier},
  author={Shibani Santurkar and Dimitris Tsipras and Brandon Tran and Andrew Ilyas and Logan Engstrom and Aleksander Madry},
  journal={ArXiv},
  year={2019},
  volume={abs/1906.09453}
}
  • Shibani Santurkar, Dimitris Tsipras, +3 authors Aleksander Madry
  • Published 2019
  • Computer Science, Mathematics
  • ArXiv
  • We show that the basic classification framework alone can be used to tackle some of the most challenging computer vision tasks. In contrast to other state-of-the-art approaches, the toolkit we develop is rather minimal: it uses a single, off-the-shelf classifier for all these tasks. The crux of our approach is that we train this classifier to be adversarially robust. It turns out that adversarial robustness is precisely what we need to directly manipulate salient features of the input. Overall… CONTINUE READING

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 17 CITATIONS

    Your Classifier is Secretly an Energy Based Model and You Should Treat it Like One

    VIEW 4 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    Proper measure for adversarial robustness

    VIEW 1 EXCERPT
    CITES METHODS

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 47 REFERENCES

    Improved Techniques for Training GANs

    Image-to-Image Translation with Conditional Adversarial Networks

    Large Scale GAN Training for High Fidelity Natural Image Synthesis

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL