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 2210.07346cs.CRcs.CVcs.LG
收藏

针对自监督学习的一种极其简单的后门攻击

An Embarrassingly Simple Backdoor Attack on Self-supervised Learning

Zhejiang University(浙江大学);Pennsylvania State University(宾夕法尼亚州立大学);Nanjing University(南京大学)

Changjiang Li, Ren Pang, Zhaohan Xi, Tianyu Du, Shouling Ji, Yuan Yao, Ting Wang

AI总结:

本文提出简单高效的自监督后门攻击 CTRL,仅需污染不超过 1% 训练数据即可实现不低于 99% 的触发误分类,并揭示 SSL 表示不变性带来的内在脆弱性。

摘要:

作为机器学习中的一种新范式,自监督学习(SSL)能够在不依赖标签的情况下学习复杂数据的高质量表示。除了消除对标注数据的需求之外,研究还发现 SSL 相比监督学习提升了对抗鲁棒性,因为缺少标签使对手更难操纵模型预测。然而,这种鲁棒性优势在多大程度上能够推广到其他类型的攻击,仍是一个悬而未决的问题。我们在后门攻击的背景下探讨这一问题。具体而言,我们设计并评估了 CTRL,这是一种极其简单却非常有效的自监督后门攻击。通过仅用难以区分的投毒样本污染极小一部分训练数据(<= 1%),CTRL 会在推理时使任何嵌入触发器的输入以高概率(>= 99%)被错误分类到对手指定的类别。我们的研究结果表明,SSL 与监督学习在面对后门攻击时同样脆弱。更重要的是,通过 CTRL 这一视角,我们研究了 SSL 对后门攻击的内在脆弱性。结合实证与分析证据,我们揭示了 SSL 中有利于对抗鲁棒性的表示不变性,可能也正是使 SSL 极易受到后门攻击影响的原因。我们的发现还意味着,现有的针对监督式后门攻击的防御措施并不容易改造,以应对 SSL 特有的脆弱性。

英文摘要:

As a new paradigm in machine learning, self-supervised learning (SSL) is capable of learning high-quality representations of complex data without relying on labels. In addition to eliminating the need for labeled data, research has found that SSL improves the adversarial robustness over supervised learning since lacking labels makes it more challenging for adversaries to manipulate model predictions. However, the extent to which this robustness superiority generalizes to other types of attacks remains an open question. We explore this question in the context of backdoor attacks. Specifically, we design and evaluate CTRL, an embarrassingly simple yet highly effective self-supervised backdoor attack. By only polluting a tiny fraction of training data (<= 1%) with indistinguishable poisoning samples, CTRL causes any trigger-embedded input to be misclassified to the adversary's designated class with a high probability (>= 99%) at inference time. Our findings suggest that SSL and supervised learning are comparably vulnerable to backdoor attacks. More importantly, through the lens of CTRL, we study the inherent vulnerability of SSL to backdoor attacks. With both empirical and analytical evidence, we reveal that the representation invariance property of SSL, which benefits adversarial robustness, may also be the very reason making \ssl highly susceptible to backdoor attacks. Our findings also imply that the existing defenses against supervised backdoor attacks are not easily retrofitted to the unique vulnerability of SSL.

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。

↑