arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2406.03404cs.LGcs.AIcs.CR
收藏

ST-DPGAN:一个用于时空数据生成的隐私保护框架

ST-DPGAN: A Privacy-preserving Framework for Spatiotemporal Data Generation

Data61, CSIRO(Data61联邦科学与工业研究组织(澳大利亚));University of Rochester(罗切斯特大学);Australian National University(澳大利亚国立大学);Tsinghua University(清华大学);RMIT University(皇家墨尔本理工大学);Qatar Mobility Innovations Center(卡塔尔交通创新中心);University of New South Wales(新南威尔士大学)

Wei Shao, Rongyi Zhu, Cai Yang, Chandra Thapa, Muhammad Ejaz Ahmed, Seyit Camtepe, Rui Zhang, DuYong Kim, Hamid Menouar, Flora D. Salim

AI总结:

针对时空数据包含敏感信息不适合第三方访问的问题,提出基于 Graph-GAN 的隐私保护生成模型,通过引入时空注意力块和反卷积结构实现 differential privacy,在保证隐私的同时维持了数据效用和预测性能。

摘要:

时空数据广泛存在于各类边缘设备中,例如用于个人通信和金融交易的设备。最近的进展激发了将时空分析与大规模语言模型相结合的日益浓厚的兴趣。然而,时空数据通常包含敏感信息,使其不适合开放的第三方访问。为了应对这一挑战,我们提出了一种基于 Graph-GAN 的模型,用于生成受隐私保护的时空数据。我们的方法在判别器中引入了空间和时间注意力块,并在生成器中采用了时空反卷积结构。这些增强使得模型能够在 Gaussian noise 下进行高效训练,从而实现 differential privacy。在三个真实的时空数据集上进行的大量实验验证了我们模型的有效性。我们的方法在保持数据效用的同时提供了隐私保证。与在原始数据上训练的模型相比,在我们生成的数据上训练的预测模型保持了具有竞争力的性能。

英文摘要:

Spatiotemporal data is prevalent in a wide range of edge devices, such as those used in personal communication and financial transactions. Recent advancements have sparked a growing interest in integrating spatiotemporal analysis with large-scale language models. However, spatiotemporal data often contains sensitive information, making it unsuitable for open third-party access. To address this challenge, we propose a Graph-GAN-based model for generating privacy-protected spatiotemporal data. Our approach incorporates spatial and temporal attention blocks in the discriminator and a spatiotemporal deconvolution structure in the generator. These enhancements enable efficient training under Gaussian noise to achieve differential privacy. Extensive experiments conducted on three real-world spatiotemporal datasets validate the efficacy of our model. Our method provides a privacy guarantee while maintaining the data utility. The prediction model trained on our generated data maintains a competitive performance compared to the model trained on the original data.

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。

↑