论文标题
关于成功和简单性：第二次查看可转移的目标攻击
On Success and Simplicity: A Second Look at Transferable Targeted Attacks
论文作者
论文摘要
据称，实现目标攻击的可转移性非常困难。当前，最新的方法是资源密集型的，因为它们需要为每个目标类别类带有其他数据的培训模型。但是，在我们的调查中，我们发现，不需要其他数据和模型训练的简单可转移攻击可以达到令人惊讶的高针对性转移性。到目前为止，这种见解一直被忽视，这主要是由于不合理地限制了攻击优化到有限数量的迭代的广泛实践。特别是，我们第一次确定简单的logit损失可以通过艺术状态产生竞争结果。我们的分析涵盖了各种传输设置，尤其是包括三个新的，现实的设置：一个没有模型相似性的合奏转移设置，较差的案例设置，具有低排名的目标类别，以及对Google Cloud Cloud Vision API的现实攻击。这些新设置中的结果表明，常见，简单的设置无法完全揭示不同攻击的实际属性，并可能引起误导性比较。我们还展示了简单的logit损失对以无数据和无培训方式生成目标通用对抗扰动的有用性。总体而言，我们的分析目的是激发对目标可转移性的更有意义的评估。代码可从https://github.com/zhengyuzhao/targeted-tansfer获得
Achieving transferability of targeted attacks is reputed to be remarkably difficult. Currently, state-of-the-art approaches are resource-intensive because they necessitate training model(s) for each target class with additional data. In our investigation, we find, however, that simple transferable attacks which require neither additional data nor model training can achieve surprisingly high targeted transferability. This insight has been overlooked until now, mainly due to the widespread practice of unreasonably restricting attack optimization to a limited number of iterations. In particular, we, for the first time, identify that a simple logit loss can yield competitive results with the state of the arts. Our analysis spans a variety of transfer settings, especially including three new, realistic settings: an ensemble transfer setting with little model similarity, a worse-case setting with low-ranked target classes, and also a real-world attack against the Google Cloud Vision API. Results in these new settings demonstrate that the commonly adopted, easy settings cannot fully reveal the actual properties of different attacks and may cause misleading comparisons. We also show the usefulness of the simple logit loss for generating targeted universal adversarial perturbations in a data-free and training-free manner. Overall, the aim of our analysis is to inspire a more meaningful evaluation on targeted transferability. Code is available at https://github.com/ZhengyuZhao/Targeted-Tansfer
