arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1901.03037cs.CVcs.LG
收藏

图像变换可以使神经网络对对抗性样本更具鲁棒性

Image Transformation can make Neural Networks more robust against Adversarial Examples

Institute of Information Security(信息安全研究所)

Dang Duy Thang, Toshihiro Matsui

摘要:

神经网络正被应用于与物联网相关的许多任务中,并取得了令人鼓舞的成果。例如,神经网络可以通过监控摄像头精确地检测人、物体和动物,以达到安全目的。然而,最近发现神经网络容易受到被称为对抗性样本的精心设计的输入样本的攻击。这样的问题会导致神经网络对人类无法感知的对抗性样本进行错误分类。我们发现,对对抗性示例图像进行旋转可以击败对抗性示例的效果。以MNIST数字图像为原始图像,首先生成对抗性样本给神经网络识别器,完全被伪造的样本欺骗。然后,我们旋转了对抗图像,并将它们交给识别器,让识别器重新获得正确的识别。因此,我们经验性地证实了对图像的旋转可以保护基于神经网络的模式识别器免受对抗性示例攻击。

英文摘要:

Neural networks are being applied in many tasks related to IoT with encouraging results. For example, neural networks can precisely detect human, objects and animal via surveillance camera for security purpose. However, neural networks have been recently found vulnerable to well-designed input samples that called adversarial examples. Such issue causes neural networks to misclassify adversarial examples that are imperceptible to humans. We found giving a rotation to an adversarial example image can defeat the effect of adversarial examples. Using MNIST number images as the original images, we first generated adversarial examples to neural network recognizer, which was completely fooled by the forged examples. Then we rotated the adversarial image and gave them to the recognizer to find the recognizer to regain the correct recognition. Thus, we empirically confirmed rotation to images can protect pattern recognizer based on neural networks from adversarial example attacks.

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。

↑