arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2205.06992cs.CRcs.LG
收藏

神经网络抗后门攻击的研究

Verifying Neural Networks Against Backdoor Attacks

Singapore Management University(新加坡管理大学)

Long H. Pham, Jun Sun

摘要:

神经网络在解决许多问题方面已经取得了最先进的性能,包括在安全/安全关键系统中的许多应用。研究人员还发现了与神经网络相关的多个安全问题。其中之一是后门攻击,即,神经网络可以嵌入后门,使得在存在触发的情况下几乎总是生成目标输出。现有的防御方法主要集中在基于语义学检测神经网络是否被“后门”,例如,激活模式据我们所知,唯一能证明不存在后门的工作是基于随机平滑,这会显著降低神经网络的性能。在这项工作中,我们提出了一种方法来验证是否一个给定的神经网络是免费的后门具有一定的成功率。我们的方法集成了统计抽样以及抽象的解释。实验结果表明,该方法可以有效地验证后门的存在或生成后门触发器。

英文摘要:

Neural networks have achieved state-of-the-art performance in solving many problems, including many applications in safety/security-critical systems. Researchers also discovered multiple security issues associated with neural networks. One of them is backdoor attacks, i.e., a neural network may be embedded with a backdoor such that a target output is almost always generated in the presence of a trigger. Existing defense approaches mostly focus on detecting whether a neural network is 'backdoored' based on heuristics, e.g., activation patterns. To the best of our knowledge, the only line of work which certifies the absence of backdoor is based on randomized smoothing, which is known to significantly reduce neural network performance. In this work, we propose an approach to verify whether a given neural network is free of backdoor with a certain level of success rate. Our approach integrates statistical sampling as well as abstract interpretation. The experiment results show that our approach effectively verifies the absence of backdoor or generates backdoor triggers.

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。

↑