Pixeldefend: Leveraging Generative Models

@inproceedings{EXAMPLES2018PixeldefendLG,
  title={Pixeldefend: Leveraging Generative Models},
  author={ADVERSARIAL EXAMPLES and Yang Song and Stefano Ermon},
  year={2018}
}
Adversarial perturbations of normal images are usually imperceptible to humans, but they can seriously confuse state-of-the-art machine learning models. What makes them so special in the eyes of image classifiers? In this paper, we show empirically that adversarial examples mainly lie in the low probability regions of the training distribution, regardless… CONTINUE READING