ImageNet-Patch：一个用于基准测试机器学习对抗补丁鲁棒性的数据集
ImageNet-Patch: A Dataset for Benchmarking Machine Learning Robustness against Adversarial Patches
AI总结：
针对对抗补丁优化成本高、鲁棒性评估易不理想的问题，提出可跨模型泛化的ImageNet-Patch数据集，经仿射变换后可快速评估鲁棒性，已在127个模型上验证有效性并开源。
摘要：
对抗补丁是输入图像中经过优化的连续像素块，会导致机器学习模型对图像进行错误分类。然而，这类补丁的优化计算成本高昂，且需要仔细调整超参数，可能会导致鲁棒性评估结果不够理想。为解决这些问题，我们提出了ImageNet-Patch，一个用于基准测试机器学习模型对抗补丁鲁棒性的数据集。该数据集包含一组经过优化、可在不同模型间泛化的补丁，经仿射变换预处理后可直接应用于ImageNet数据。这一流程借助对抗扰动的可迁移性，可实现近似但更快速的鲁棒性评估。我们通过测试所生成补丁对127个模型的攻击效果，展示了该数据集的实用性。最后，我们讨论了该数据集如何作为鲁棒性基准使用，以及我们的方法如何推广到其他领域。我们已在https://github.com/pralab/ImageNet-Patch开源了数据集与评估代码。
英文摘要：
Adversarial patches are optimized contiguous pixel blocks in an input image that cause a machine-learning model to misclassify it. However, their optimization is computationally demanding, and requires careful hyperparameter tuning, potentially leading to suboptimal robustness evaluations. To overcome these issues, we propose ImageNet-Patch, a dataset to benchmark machine-learning models against adversarial patches. It consists of a set of patches, optimized to generalize across different models, and readily applicable to ImageNet data after preprocessing them with affine transformations. This process enables an approximate yet faster robustness evaluation, leveraging the transferability of adversarial perturbations. We showcase the usefulness of this dataset by testing the effectiveness of the computed patches against 127 models. We conclude by discussing how our dataset could be used as a benchmark for robustness, and how our methodology can be generalized to other domains. We open source our dataset and evaluation code at https://github.com/pralab/ImageNet-Patch.
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。
