arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 1809.09262stat.MLcs.CRcs.LGcs.NE
收藏

具有对抗性攻击结构抵抗力的神经网络

Neural Networks with Structural Resistance to Adversarial Attacks

University of California, Santa Cruz(加州大学圣克鲁兹分校)

Luca de Alfaro

摘要:

在对机器学习分类器的对抗性攻击中,小扰动被添加到正确分类的输入中。扰动产生了对抗性的例子,这些例子与未扰动的输入几乎无法区分,但却被错误分类。在用于深度学习的标准神经网络中,攻击者可以从大多数输入中制作对抗性示例,以导致他们选择的错误分类。我们引入了一种新型的网络单元,称为RBFI单元,其非线性结构使其具有固有的抵抗对抗性攻击的能力。在置换不变的MNIST上,在没有对抗性攻击的情况下,使用RBFI单元的网络与使用sigmoid单元的网络的性能相匹配,并且略低于使用ReLU单元的网络的准确性。当受到对抗性攻击时,具有RBFI单元的网络保持90%以上的准确率,而具有ReLU或sigmoid单元的网络的准确率降低到2%以下。使用常规输入训练的RBFI网络在抵抗对抗性攻击方面甚至优于在对抗性示例帮助下训练的ReLU和sigmoid网络。RBFI单元的非线性结构使得它们难以使用标准梯度下降来训练。我们表明,RBFI单元的网络可以使用伪梯度有效地训练到高精度,使用专门设计的函数来计算,以方便学习,而不是它们的真实导数。我们证明了伪梯度的使用使得训练深度RBFI网络变得实用,并且我们比较了RBFI网络的几种结构选择的准确性。

英文摘要:

In adversarial attacks to machine-learning classifiers, small perturbations are added to input that is correctly classified. The perturbations yield adversarial examples, which are virtually indistinguishable from the unperturbed input, and yet are misclassified. In standard neural networks used for deep learning, attackers can craft adversarial examples from most input to cause a misclassification of their choice. We introduce a new type of network units, called RBFI units, whose non-linear structure makes them inherently resistant to adversarial attacks. On permutation-invariant MNIST, in absence of adversarial attacks, networks using RBFI units match the performance of networks using sigmoid units, and are slightly below the accuracy of networks with ReLU units. When subjected to adversarial attacks, networks with RBFI units retain accuracies above 90% for attacks that degrade the accuracy of networks with ReLU or sigmoid units to below 2%. RBFI networks trained with regular input are superior in their resistance to adversarial attacks even to ReLU and sigmoid networks trained with the help of adversarial examples. The non-linear structure of RBFI units makes them difficult to train using standard gradient descent. We show that networks of RBFI units can be efficiently trained to high accuracies using pseudogradients, computed using functions especially crafted to facilitate learning instead of their true derivatives. We show that the use of pseudogradients makes training deep RBFI networks practical, and we compare several structural alternatives of RBFI networks for their accuracy.

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。

↑