SaTC: CORE: Small: Collaborative: When Adversarial Learning Meets Differential Privacy: Theoretical Foundation and Applications

SaTC：核心：小型：协作：当对抗性学习遇到差异性隐私时：理论基础和应用

基本信息

批准号：
1935923
负责人：
My Thai
金额：
$ 25万
依托单位：
University of Florida
依托单位国家：
美国
项目类别：
Standard Grant
财政年份：
2020
资助国家：
美国
起止时间：
2020-07-01 至 2024-06-30
项目状态：
已结题

来源：
https://www.nsf.gov/awardsearch/showAward?AWD_ID=1935923&HistoricalAwards=false
关键词：
SaTC CORE Small Collaborative When

项目摘要

The pervasiveness of machine learning exposes new and severe vulnerabilities in software systems, where deployed deep neural networks can be exploited to reveal sensitive information in private training data, and to make the models misclassify. However, existing learning algorithms have not been designed to be simultaneously robust to such privacy and integrity attacks, in both theory and practice. In field trials, such lack of protection and efficacy significantly degrades the performance of machine learning-based systems, and puts sensitive data at high risk, thereby exposing service providers to legal action based on HIPAA/HITECH law and related regulations. This project aims to develop the first framework to advance and seamlessly integrate key techniques, including adversarial learning, privacy preserving, and certified defenses, offering tight and reliable protection against both privacy and integrity attacks, while retaining high model utility in deep neural networks. The system is being developed for scalable, complex, and commonly used machine learning frameworks, providing a fundamental impact to both industry and educational environments.An ultimate goal of this project is to build a core foundation of privacy preservation in adversarial learning, to better address the trade-off between model utility, privacy loss, and certified defenses. Accordingly, the team theoretically connects adversarial learning and privacy preservation by introducing a new set of rigorous theories to address the trade-off between model utility and privacy loss. To further strengthen the safety of the system, the team will conduct a new class of attacks towards discovering previously unknown and unprotected vulnerabilities, including highly sensitive and hidden correlation structures among data instances, which will be used to amplify existing model attacks. Based upon that effort, vulnerable features and correlations will be automatically identified and protected, towards unified robust and privacy preserving learning, given both model training and inference. Finally, the team will optimize the trade-off among model utility, privacy loss, and certified defenses. The project is expected to lay a theoretical and practical foundation of key privacy-preserving techniques to protect users' personal and highly sensitive data in adversarial learning under model attacks.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

机器学习的普遍性暴露了软件系统中的新脆弱性，可以利用部署的深层神经网络在私人培训数据中揭示敏感信息，并使模型错误分类。但是，在理论和实践中，现有的学习算法并未设计为对这种隐私和诚信攻击的同时稳健。在现场试验中，这种缺乏保护和功效会大大降低基于机器学习的系统的性能，并使敏感数据处于高风险，从而使服务提供商根据HIPAA/HITECH/HITECH LAW及其相关法规将服务提供商暴露于法律诉讼中。该项目旨在开发第一个框架，以促进和无缝整合关键技术，包括对抗性学习，保留隐私和认证的防御，从而对隐私和诚信攻击提供严格而可靠的保护，同时保留深层神经网络中的高模型效用。该系统是针对可扩展，复杂且常用的机器学习框架开发的，对行业和教育环境产生了根本的影响。该项目的最终目标是在对抗性学习中建立隐私保护的核心基础，以更好地解决问题模型效用，隐私损失和认证防御能力之间的权衡。因此，从理论上讲，团队通过引入一套新的严格理论来解决模型效用与隐私损失之间的权衡，从而将对抗性学习和隐私保护联系起来。为了进一步加强系统的安全性，团队将进行新的攻击，以发现以前未知和未受保护的漏洞，包括数据实例之间高度敏感和隐藏的相关结构，这些结构将用于扩大现有的模型攻击。基于这一努力，鉴于模型培训和推理，将自动识别和保护脆弱的特征和相关性，以朝着统一的稳健和隐私保留学习。最后，团队将优化模型公用事业，隐私损失和认证防御的权衡。预计该项目将在模型攻击下为用户在对抗性学习中保护用户的个人和高度敏感数据的主要隐私技术的理论和实用基础。这项奖项反映了NSF的法定任务，并被认为是通过使用评估来支持的。基金会的智力优点和更广泛的影响评论标准。

项目成果

期刊论文数量（8）

专著数量（0）

科研奖励数量（0）

会议论文数量（0）

专利数量（0）

XRand: Differentially Private Defense against Explanation-Guided Attacks

DOI：
10.48550/arxiv.2212.04454
发表时间：
2022-12
期刊：
影响因子：
0
作者：
Truc D. T. Nguyen;Phung Lai;Nhathai Phan;M. Thai
通讯作者：
Truc D. T. Nguyen;Phung Lai;Nhathai Phan;M. Thai

Continual Learning with Differential Privacy

具有差异隐私的持续学习

DOI：
10.1007/978-3-030-92310-5_39
发表时间：
2021
期刊：
International Conference on Neural Information Processing
影响因子：
0
作者：
Desai, Pradnya;Lai, Phung;Phan, NhatHai;Thai, My T.
通讯作者：
Thai, My T.

Scalable Differential Privacy with Certified Robustness in Adversarial Learning

DOI：
发表时间：
2019-03
期刊：
影响因子：
0
作者：
HaiNhat Phan;M. Thai;Han Hu;R. Jin;Tong Sun;D. Dou
通讯作者：
HaiNhat Phan;M. Thai;Han Hu;R. Jin;Tong Sun;D. Dou

Active Membership Inference Attack under Local Differential Privacy in Federated Learning

DOI：
10.48550/arxiv.2302.12685
发表时间：
2023-02
期刊：
ArXiv
影响因子：
0
作者：
Truc D. T. Nguyen;Phung Lai;K. Tran;Nhathai Phan;M. Thai
通讯作者：
Truc D. T. Nguyen;Phung Lai;K. Tran;Nhathai Phan;M. Thai

Heterogeneous Randomized Response for Differential Privacy in Graph Neural Network

图神经网络中差分隐私的异构随机响应

DOI：
发表时间：
2022
期刊：
IEEE Big Data
影响因子：
0
作者：
Tran, Khang;Lai, Phung;Phan, NhatHai;Khalil, Issa;Ma, Yao;Khreishah, Abdallah;Thai, My T;Wu, Xintao
通讯作者：
Wu, Xintao