TY - UNPB
T1 - Patch augmentation: Towards efficient decision boundaries for neural networks
AU - Bloice, Marcus Daniel
AU - Roth, Peter M.
AU - Holzinger, Andreas
PY - 2019/11/8
Y1 - 2019/11/8
N2 - In this paper we propose a new augmentation technique, called patch augmentation, that, in our experiments, improves model accuracy and makes networks more robust to adversarial attacks. In brief, this data-independent approach creates new image data based on image/label pairs, where a patch from one of the two images in the pair is superimposed on to the other image, creating a new augmented sample. The new image's label is a linear combination of the image pair's corresponding labels. Initial experiments show a several percentage point increase in accuracy on CIFAR-10, from a baseline of approximately 81% to 89%. CIFAR-100 sees larger improvements still, from a baseline of 52% to 68% accuracy. Networks trained using patch augmentation are also more robust to adversarial attacks, which we demonstrate using the Fast Gradient Sign Method.
AB - In this paper we propose a new augmentation technique, called patch augmentation, that, in our experiments, improves model accuracy and makes networks more robust to adversarial attacks. In brief, this data-independent approach creates new image data based on image/label pairs, where a patch from one of the two images in the pair is superimposed on to the other image, creating a new augmented sample. The new image's label is a linear combination of the image pair's corresponding labels. Initial experiments show a several percentage point increase in accuracy on CIFAR-10, from a baseline of approximately 81% to 89%. CIFAR-100 sees larger improvements still, from a baseline of 52% to 68% accuracy. Networks trained using patch augmentation are also more robust to adversarial attacks, which we demonstrate using the Fast Gradient Sign Method.
U2 - arXiv preprint arXiv:1911.07922
DO - arXiv preprint arXiv:1911.07922
M3 - Preprint
T3 - arXiv.org e-Print archive
BT - Patch augmentation: Towards efficient decision boundaries for neural networks
ER -