Patch augmentation: Towards efficient decision boundaries for neural networks

Authors: Marcus D. Bloice, Peter M. Roth, Andreas Holzinger
Published: 2019-11-08 | Updated: 2019-11-25

Source: https://arxiv.org/abs/1911.07922

PDF: https://arxiv.org/pdf/1911.07922

Labels Predicted by AI

モデル性能評価 敵対的攻撃手法 特徴エンジニアリング

Please note that these labels were automatically added by AI. Therefore, they may not be entirely accurate.
For more details, please see the About the Literature Database page.

Abstract

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 baseline of 52 also more robust to adversarial attacks, which we demonstrate using the Fast Gradient Sign Method.