The exploitation of publicly accessible data has led to escalating concerns regarding data privacy and intellectual property (IP) breaches in the age of artificial intelligence. To safeguard both data privacy and IP-related domain knowledge, efforts have been undertaken to render shared data unlearnable for unauthorized models in the wild. Existing methods apply empirically optimized perturbations to the data in the hope of disrupting the correlation between the inputs and the corresponding labels such that the data samples are converted into Unlearnable Examples (UEs). Nevertheless, the absence of mechanisms to verify the robustness of UEs against uncertainty in unauthorized models and their training procedures engenders several under-explored challenges. First, it is hard to quantify the unlearnability of UEs against unauthorized adversaries from different runs of training, leaving the soundness of the defense in obscurity. Particularly, as a prevailing evaluation metric, empirical test accuracy faces generalization errors and may not plausibly represent the quality of UEs. This also leaves room for attackers, as there is no rigid guarantee of the maximal test accuracy achievable by attackers. Furthermore, we find that a simple recovery attack can restore the clean-task performance of the classifiers trained on UEs by slightly perturbing the learned weights. To mitigate the aforementioned problems, in this paper, we propose a mechanism for certifying the so-called $(q, η)$-Learnability of an unlearnable dataset via parametric smoothing. A lower certified $(q, η)$-Learnability indicates a more robust and effective protection over the dataset. Concretely, we 1) improve the tightness of certified $(q, η)$-Learnability and 2) design Provably Unlearnable Examples (PUEs) which have reduced $(q, η)$-Learnability.
Academy of Mathematics and Systems Science, Chinese Academy of Sciences（中国科学院数学与系统科学研究院）；Institute of Software, Chinese Academy of Sciences（中国科学院软件研究所）；University of Chinese Academy of Sciences（中国科学院大学）
Learning the Unlearnable: Adversarial Augmentations Suppress Unlearnable Example Attacks
Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences（中国科学院深圳先进技术研究院）；University of Chinese Academy of Sciences（中国科学院大学）；University of Macau（澳门大学）
Channel-Level Semantic Perturbations: Unlearnable Examples for Diverse Training Paradigms
School of Information and Communication Engineering, Dalian University of Technology（大连理工大学信息与通信工程学院）；School of Computer Science and Technology, Anhui University（安徽大学计算机科学与技术学院）；School of Computer Science and Technology, Zhejiang University（浙江大学计算机科学与技术学院）
Why Do Unlearnable Examples Work: A Novel Perspective of Mutual Information
State Key Laboratory of Mathematical Sciences, Academy of Mathematics and Systems Science, Chinese Academy of Sciences（数学科学国家重点实验室，数学与系统科学研究院，中国科学院）；University of Chinese Academy of Sciences（中国科学院大学）；College of AI, Tsinghua University（清华大学人工智能学院）；Shanghai Qi Zhi Institute（上海启智研究院）
Unlearnable Examples: Making Personal Data Unexploitable
The University of Melbourne（墨尔本大学）；Deakin University（迪肯大学）；Peking University（北京大学）
互联网上大量的“免费”数据一直是深度学习目前成功的关键。然而，它也引发了对未经授权利用个人数据来训练商业模型的隐私担忧。因此，至关重要的是要制定方法，以防止未经授权的数据利用。本文提出了一个问题：\emph{can data be made unlearnable for deep learning models?}我们提出了一种类型的\emph{error-minimizing}噪音，确实可以使训练样本不可学习。有意生成误差最小化噪声，以将一个或多个训练样本的误差减小到接近零，这可以欺骗模型，使其相信从这些样本中“没有什么”可以学习。噪声被限制为人眼无法察觉，因此不会影响正常的数据效用。我们经验验证的有效性，误差最小化的噪音在样本明智和类明智的形式。我们还展示了其灵活性下广泛的实验设置和实用性的人脸识别的案例研究。我们的工作为使个人数据无法被深度学习模型利用迈出了重要的第一步。
Faculty of Engineering and Information Technology, The University of Melbourne（墨尔本大学工程与信息技术学院）；School of Computer Science, Fudan University（复旦大学计算机科学学院）
ARMOR: Shielding Unlearnable Examples against Data Augmentation
Nanyang Technological University（南洋理工大学）；College of Electrical Engineering, Zhejiang University（浙江大学电气工程学院）；School of Cyber Science and Engineering, Wuhan University（武汉大学网络科学与工程学院）；ZJU-Hangzhou Global Scientific and Technological Innovation Center (HIC)（浙江大学杭州国际科技创新中心（HIC））；State Key Laboratory of Blockchain and Data Security, Zhejiang University（浙江大学区块链与数据安全国家重点实验室）
UTOPIA: Unlearnable Tabular Data via Decoupled Shortcut Embedding
University of Electronic Science and Technology of China（电子科学与技术大学）；Independent Researcher（独立研究者）；University of Science and Technology of China（科学技术大学）；Nanyang Technological University（南洋理工大学）
Provably efficient variational generative modeling of quantum many-body systems via quantum-probabilistic information geometry
Google Brain（谷歌大脑）；Institute for Quantum Computing, University of Waterloo（滑铁卢大学量子计算研究所）；UC Berkeley（加州大学伯克利分校）；Stanford University（斯坦福大学）；Princeton University（普林斯顿大学）；University of Waterloo（滑铁卢大学）
Model Mimic Attack: Knowledge Distillation for Provably Transferable Adversarial Examples
Research Center for Trusted Artificial Intelligence（可信人工智能研究中心）；Ivannikov Institute for System Programming of the Russian Academy of Sciences（俄罗斯科学院伊万尼科夫系统编程研究所）；Moscow Institute of Physics and Technology (National Research University)（莫斯科物理技术学院（国立研究型大学））；AIRI（AIRI（人工智能研究所））
NSF AI Institute for Artificial Intelligence and Fundamental Interactions（NSF人工智能与基本相互作用人工智能研究所）；Laboratory for Nuclear Science, MIT（麻省理工学院核科学实验室）；MIT（麻省理工学院）
