本文探讨深度神经网络中损失关于输入数据的曲率性质。损失关于输入的曲率(称为输入损失曲率)是损失关于输入的 Hessian 矩阵的迹。我们研究输入损失曲率在训练集和测试集之间的变化,及其对训练-测试可区分性的影响。我们构建了一个理论框架,基于隐私和训练集大小推导出训练-测试可区分性的上界。这一新见解推动了利用输入损失曲率的新型黑盒成员推理攻击的发展。我们通过在计算机视觉分类任务中的实验验证了理论发现,证明输入损失曲率在成员推理有效性上超越了现有方法。我们的分析揭示了成员推理攻击(MIA)方法的性能如何随训练集大小变化,表明基于曲率的 MIA 在足够大的数据集上优于其他方法。这一条件通常被真实数据集满足,正如我们在 CIFAR10、CIFAR100 和 ImageNet 上的结果所示。这些发现不仅增进了我们对深度神经网络行为的理解,还提升了测试机器学习中隐私保护技术的能力。
英文摘要:
In this paper, we explore the properties of loss curvature with respect to input data in deep neural networks. Curvature of loss with respect to input (termed input loss curvature) is the trace of the Hessian of the loss with respect to the input. We investigate how input loss curvature varies between train and test sets, and its implications for train-test distinguishability. We develop a theoretical framework that derives an upper bound on the train-test distinguishability based on privacy and the size of the training set. This novel insight fuels the development of a new black box membership inference attack utilizing input loss curvature. We validate our theoretical findings through experiments in computer vision classification tasks, demonstrating that input loss curvature surpasses existing methods in membership inference effectiveness. Our analysis highlights how the performance of membership inference attack (MIA) methods varies with the size of the training set, showing that curvature-based MIA outperforms other methods on sufficiently large datasets. This condition is often met by real datasets, as demonstrated by our results on CIFAR10, CIFAR100, and ImageNet. These findings not only advance our understanding of deep neural network behavior but also improve the ability to test privacy-preserving techniques in machine learning.
PAT: Privacy-Preserving Adversarial Transfer for Accurate, Robust and Privacy-Preserving EEG Decoding
Hubei Key Laboratory of Brain-inspired Intelligent Systems, School of Artificial Intelligence and Automation, Huazhong University of Science and Technology(华中科技大学人工智能与自动化学院脑启发智能系统湖北省重点实验室);Shenzhen Huazhong University of Science and Technology Research Institute(深圳华中科技大学研究院);Zhongguancun Academy(中关村学院)
SAFE: Secure and Accurate Federated Learning for Privacy-Preserving Brain-Computer Interfaces
Key Laboratory of the Ministry of Education for Image Processing and Intelligent Control, School of Artificial Intelligence and Automation, Huazhong University of Science and Technology(教育部图像处理与智能控制重点实验室,人工智能与自动化学院,华中科技大学);Zhongguancun Academy(中关村学院)
Tailoring deep learning for real-time brain-computer interfaces: From offline models to calibration-free online decoding
Institute of Signal Processing and System Theory, University of Stuttgart(信号处理与系统理论研究所,斯图加特大学);Clinical Neurotechnology Laboratory, Department of Psychiatry and Neurosciences, Charité Campus Mitte (CCM), Charité - Universitätsmedizin Berlin(临床神经技术实验室,精神病学与神经科学系,Charité-柏林大学医学院)
Fraunhofer Heinrich Hertz Institute(弗劳恩霍夫海因里希·赫兹研究所);Technical University of Berlin(柏林工业大学);Berlin Institute for the Foundations of Learning and Data (BIFOLD)(柏林学习与数据基础研究所(BIFOLD))
Deep Neural Networks and Brain Alignment: Brain Encoding and Decoding (Survey)
Inria Bordeaux(法国国家信息与自动化研究所波尔多分所);LaBRI, Université de Bordeaux(波尔多大学拉布里实验室);Université de Bordeaux(波尔多大学);TU Berlin(柏林工业大学);National University of Singapore(新加坡国立大学);Microsoft(微软公司);International Institute of Information Technology Hyderabad(海得拉巴国际信息技术研究所);CEA(法国原子能和替代能源委员会);CNRS(法国国家科学研究中心)