Indiscriminate data poisoning attacks are quite effective against supervised learning. However, not much is known about their impact on unsupervised contrastive learning (CL). This paper is the first to consider indiscriminate poisoning attacks of contrastive learning. We propose Contrastive Poisoning (CP), the first effective such attack on CL. We empirically show that Contrastive Poisoning, not only drastically reduces the performance of CL algorithms, but also attacks supervised learning models, making it the most generalizable indiscriminate poisoning attack. We also show that CL algorithms with a momentum encoder are more robust to indiscriminate poisoning, and propose a new countermeasure based on matrix completion. Code is available at: https://github.com/kaiwenzha/contrastive-poisoning.

翻译：无差别数据投毒攻击在监督学习中效果显著，但对其在无监督对比学习（CL）中的影响尚不明确。本文首次探讨了对比学习中的无差别投毒攻击，提出了对比投毒（Contrastive Poisoning, CP）方法——这是首个针对CL的有效无差别投毒攻击。实验证明，对比投毒不仅能显著降低CL算法的性能，还能攻击监督学习模型，成为最具泛化性的无差别投毒攻击。我们还发现，带有动量编码器的CL算法对无差别投毒更具鲁棒性，并提出了一种基于矩阵补全的新防御措施。代码见：https://github.com/kaiwenzha/contrastive-poisoning。