arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2005.07998cs.LGcs.CRstat.ML
收藏

基于加密技术的视觉分类对抗防御

Encryption Inspired Adversarial Defense for Visual Classification

Tokyo Metropolitan University(东京都立大学)

MaungMaung AprilPyone, Hitoshi Kiya

摘要:

无论模型是否受到攻击,传统的对抗性防御都会降低分类准确性。此外,大多数基于图像处理的防御由于模糊梯度的问题而失败。在本文中,我们提出了一种新的对抗性防御,这是一个防御性的变换的训练和测试图像的启发感知图像加密方法。所提出的方法利用了一个秘密密钥的逐块像素洗牌方法。实验进行了自适应和非自适应最大范数有界白盒攻击,同时考虑混淆梯度。实验结果表明,该方法在CIFAR-10数据集上,在噪声距离为8/255的干净图像和对抗性样本上分别达到了91.55%和89.66%的准确率。因此,所提出的防御优于最先进的对抗性防御,包括潜在对抗性训练,对抗性训练和温度计编码。

英文摘要:

Conventional adversarial defenses reduce classification accuracy whether or not a model is under attacks. Moreover, most of image processing based defenses are defeated due to the problem of obfuscated gradients. In this paper, we propose a new adversarial defense which is a defensive transform for both training and test images inspired by perceptual image encryption methods. The proposed method utilizes a block-wise pixel shuffling method with a secret key. The experiments are carried out on both adaptive and non-adaptive maximum-norm bounded white-box attacks while considering obfuscated gradients. The results show that the proposed defense achieves high accuracy (91.55 %) on clean images and (89.66 %) on adversarial examples with noise distance of 8/255 on CIFAR-10 dataset. Thus, the proposed defense outperforms state-of-the-art adversarial defenses including latent adversarial training, adversarial training and thermometer encoding.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑