Machine learning (ML) has established itself as a cornerstone for various critical applications ranging from autonomous driving to authentication systems. However, with this increasing adoption rate of machine learning models, multiple attacks have emerged. One class of such attacks is training time attack, whereby an adversary executes their attack before or during the machine learning model training. In this work, we propose a new training time attack against computer vision based machine learning models, namely model hijacking attack. The adversary aims to hijack a target model to execute a different task than its original one without the model owner noticing. Model hijacking can cause accountability and security risks since a hijacked model owner can be framed for having their model offering illegal or unethical services. Model hijacking attacks are launched in the same way as existing data poisoning attacks. However, one requirement of the model hijacking attack is to be stealthy, i.e., the data samples used to hijack the target model should look similar to the model's original training dataset. To this end, we propose two different model hijacking attacks, namely Chameleon and Adverse Chameleon, based on a novel encoder-decoder style ML model, namely the Camouflager. Our evaluation shows that both of our model hijacking attacks achieve a high attack success rate, with a negligible drop in model utility.
翻译:机器学习(ML)已经成为从自主驾驶到认证系统等各种关键应用的基石。然而,随着机器学习模式的采用率不断提高,出现了多重袭击。这类袭击的一类是培训时间攻击,即对手在机器学习模式培训之前或期间进行攻击。在这项工作中,我们提议对基于计算机视觉的机器学习模式,即劫持模型袭击进行新的培训时间攻击。对手的目的是劫持一个目标模式,以便执行不同于原目标模式的任务,而没有模型所有人注意。由于被劫持的模型所有人可以设计其提供非法或不道德服务的模型,因此可以产生问责和安全风险。典型劫持袭击的发起方式与现有数据中毒袭击相同。但是,劫持模型袭击的一个要求是隐蔽性,也就是说,劫持目标模型使用的数据样本应该类似于模型最初的培训数据集。为此,我们建议了两种不同的劫持模型袭击模式,即Chameleon和Adverse Chameleon,, 其依据一种新型的编码解码模式ML型袭击模式,即Camouglaging 成功率。我们的评估显示,我们的两个模型都是一种可忽略的通用式袭击。