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 2204.09092cs.LGcs.CR
收藏

对神经网络的无差别数据投毒攻击

Indiscriminate Data Poisoning Attacks on Neural Networks

University of Waterloo(滑铁卢大学)

Yiwei Lu, Gautam Kamath, Yaoliang Yu

AI总结:

本文将数据投毒攻击与序列Stackelberg博弈算法相联系,利用二阶信息优化和自动微分包,设计了能同时生成数万个投毒点的高效攻击方法,并实证验证了其对神经网络的有效性。

摘要:

数据投毒攻击通过在训练过程中注入“投毒”数据来影响模型,近期备受关注。本文深入分析了现有投毒攻击,并将其与求解序列Stackelberg博弈的新老算法相联系。通过为攻击者选择合适的损失函数,并利用二阶信息的算法进行优化,我们设计了对神经网络有效的投毒攻击。我们提出了利用现代自动微分包的高效实现,能够同时协调生成数万个投毒点,这与逐一生成投毒点的现有方法形成对比。我们进一步进行了大量实验,实证探索了数据投毒攻击对深度神经网络的影响。

英文摘要:

Data poisoning attacks, in which a malicious adversary aims to influence a model by injecting "poisoned" data into the training process, have attracted significant recent attention. In this work, we take a closer look at existing poisoning attacks and connect them with old and new algorithms for solving sequential Stackelberg games. By choosing an appropriate loss function for the attacker and optimizing with algorithms that exploit second-order information, we design poisoning attacks that are effective on neural networks. We present efficient implementations that exploit modern auto-differentiation packages and allow simultaneous and coordinated generation of tens of thousands of poisoned points, in contrast to existing methods that generate poisoned points one by one. We further perform extensive experiments that empirically explore the effect of data poisoning attacks on deep neural networks.

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。

↑