arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2106.04569cs.CVcs.AIcs.CYcs.LG
收藏

人脸识别模型的模拟对抗测试

Simulated Adversarial Testing of Face Recognition Models

Boston University(波士顿大学);Johns Hopkins University(约翰斯·霍普金斯大学);UC Santa Cruz(加州大学圣克鲁兹分校)

Nataniel Ruiz, Adam Kortylewski, Weichao Qiu, Cihang Xie, Sarah Adel Bargal, Alan Yuille, Stan Sclaroff

摘要:

大多数机器学习模型都是在固定的数据集上进行验证和测试的。这可能会对模型的功能和弱点给出不完整的描述。这些弱点可以在现实世界中的测试中暴露出来。这种故障所涉及的风险可能是利润损失、时间损失,甚至在某些关键应用中可能是生命损失。为了缓解这个问题,可以使用可解释的参数以细粒度的方式控制模拟器来探索语义图像流形。在这项工作中,我们提出了一个框架,用于学习如何使用模拟器以对抗的方式测试机器学习算法,以便在关键场景中部署模型之前找到模型中的弱点。我们将此方法应用于人脸识别系统中。我们表明,使用模拟样本可以发现在真实数据上训练的模型的某些弱点。使用我们提出的方法,我们可以找到欺骗当代人脸识别模型的对抗性合成人脸。这表明,这些模型具有无法通过常用验证数据集测量的弱点。我们假设这种类型的对抗性示例不是孤立的,而是通常位于模拟器的潜在空间中的连通空间中。我们提出了一种方法来找到这些对抗区域,而不是在对抗示例文献中发现的典型对抗点。

英文摘要:

Most machine learning models are validated and tested on fixed datasets. This can give an incomplete picture of the capabilities and weaknesses of the model. Such weaknesses can be revealed at test time in the real world. The risks involved in such failures can be loss of profits, loss of time or even loss of life in certain critical applications. In order to alleviate this issue, simulators can be controlled in a fine-grained manner using interpretable parameters to explore the semantic image manifold. In this work, we propose a framework for learning how to test machine learning algorithms using simulators in an adversarial manner in order to find weaknesses in the model before deploying it in critical scenarios. We apply this method in a face recognition setup. We show that certain weaknesses of models trained on real data can be discovered using simulated samples. Using our proposed method, we can find adversarial synthetic faces that fool contemporary face recognition models. This demonstrates the fact that these models have weaknesses that are not measured by commonly used validation datasets. We hypothesize that this type of adversarial examples are not isolated, but usually lie in connected spaces in the latent space of the simulator. We present a method to find these adversarial regions as opposed to the typical adversarial points found in the adversarial example literature.

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。

↑