Backdoor Attacks on Self-Supervised Learning-猫眼课题宝

DOI：

10.1109/cvpr52688.2022.01298

发表时间：

2021-05

期刊：

2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

影响因子：

通讯作者：

中科院分区：

其他

文献类型：

作者：

Aniruddha Saha;Ajinkya Tejankar;Soroush Abbasi Koohpayegani;H. Pirsiavash

大规模未标记数据刺激了学习丰富视觉表征的自监督学习方法的最新进展。最先进的从图像中学习表示的自监督方法（例如 MoCo、BYOL、MSF）使用归纳偏差，即图像的随机增强（例如随机裁剪）应该产生类似的嵌入。我们表明，此类方法容易受到后门攻击 - 攻击者通过向图像添加触发器（攻击者选择的图像补丁）来毒害一小部分未标记的数据。模型在干净的测试图像上表现良好，但攻击者可以通过在测试时显示触发器来操纵模型的决策。后门攻击在监督学习中得到了广泛的研究，据我们所知，我们是第一个将后门攻击用于自监督学习的人。后门攻击在自监督学习中更为实用，因为大量未标记数据的使用使得数据检查以消除毒物变得令人望而却步。我们表明，在我们的针对性攻击中，攻击者可以通过在测试时使用触发器对目标类别产生许多误报。我们还提出了一种基于知识蒸馏的防御方法，成功地抵消了攻击。我们的代码可以在这里找到：https://github.com/UMBCvisionISSL-Backdoor

Large-scale unlabeled data has spurred recent progress in self-supervised learning methods that learn rich vi-sual representations. State-of-the-art self-supervised methods for learning representations from images (e.g., MoCo, BYOL, MSF) use an inductive bias that random augmentations (e.g., random crops) of an image should produce similar embeddings. We show that such methods are vulnerable to backdoor attacks - where an attacker poisons a small part of the unlabeled data by adding a trigger (image patch chosen by the attacker) to the images. The model performance is good on clean test images, but the attacker can manipulate the decision of the model by showing the trigger at test time. Backdoor attacks have been studied extensively in supervised learning and to the best of our knowledge, we are the first to study them for self-supervised learning. Backdoor attacks are more practical in self-supervised learning, since the use of large unlabeled data makes data inspection to remove poisons prohibitive. We show that in our targeted attack, the attacker can produce many false positives for the target category by using the trigger at test time. We also propose a defense method based on knowledge distillation that succeeds in neutralizing the attack. Our code is available here: https://github.com/UMBCvisionISSL-Backdoor

Backdoor Attacks on Self-Supervised Learning

文献摘要

相似文献