arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2305.00645cs.CR
收藏

GTree:GPU 友好的隐私保护决策树训练与推理

GTree: GPU-Friendly Privacy-preserving Decision Tree Training and Inference

Qifan Wang, Shujie Cui, Lei Zhou, Ye Dong, Jianli Bai, Yun Sing Koh, Giovanni Russello

AI总结:

本文提出 GTree,首个利用 GPU 加速 MPC 保护的安全决策树训练与推理方案,设计了 GPU 友好的协议,在提升训练和推理效率的同时实现了更强的安全保证。

摘要:

决策树(DT)因其通用性、速度和可解释性而成为一种广泛使用的机器学习模型。然而,对于隐私敏感的应用,将 DT 训练和推理外包给云平台会引发对数据隐私的担忧。研究人员利用安全多方计算(MPC)等加密原语,开发了用于 DT 训练和推理的隐私保护方法。尽管这些方法取得了进展,但它们仍然面临繁重的计算和通信开销。最近很少有研究使用图形处理器(GPU)来提升受 MPC 保护的深度学习的性能。这自然引出了一个问题:受 MPC 保护的 DT 训练和推理能否通过 GPU 加速?我们提出了 GTree,这是首个利用 GPU 加速受 MPC 保护的安全 DT 训练和推理的方案。GTree 由 3 个参与方共同构建,它们利用 GPU 安全且联合地执行 DT 训练和推理的每一步。GTree 中的每个 MPC 协议都被设计为 GPU 友好的版本。性能评估表明,与之前最高效的基于 CPU 的工作相比,GTree 在训练 SPECT 和 Adult 数据集时分别实现了约 11 倍和约 21 倍的提升。在推理方面,当 DT 的层数少于 10 层时,GTree 展现出卓越的效率;在使用 7 层的树对 10^4 个实例进行推理时,它比之前最高效的工作快 126 倍。GTree 还实现了比以往解决方案更强的安全保证,以往解决方案会泄露树的结构,而 GTree 仅泄露树的深度和数据样本大小。借助 oblivious array access,GPU 上的访问模式也得到了保护。

英文摘要:

Decision tree (DT) is a widely used machine learning model due to its versatility, speed, and interpretability. However, for privacy-sensitive applications, outsourcing DT training and inference to cloud platforms raise concerns about data privacy. Researchers have developed privacy-preserving approaches for DT training and inference using cryptographic primitives, such as Secure Multi-Party Computation (MPC). While these approaches have shown progress, they still suffer from heavy computation and communication overheads. Few recent works employ Graphical Processing Units (GPU) to improve the performance of MPC-protected deep learning. This raises a natural question: \textit{can MPC-protected DT training and inference be accelerated by GPU?} We present GTree, the first scheme that uses GPU to accelerate MPC-protected secure DT training and inference. GTree is built across 3 parties who securely and jointly perform each step of DT training and inference with GPU. Each MPC protocol in GTree is designed in a GPU-friendly version. The performance evaluation shows that GTree achieves ${\thicksim}11{\times}$ and ${\thicksim}21{\times}$ improvements in training SPECT and Adult datasets, compared to the prior most efficient CPU-based work. For inference, GTree shows its superior efficiency when the DT has less than 10 levels, which is $126\times$ faster than the prior most efficient work when inferring $10^4$ instances with a tree of 7 levels. GTree also achieves a stronger security guarantee than prior solutions, which only leaks the tree depth and size of data samples while prior solutions also leak the tree structure. With \textit{oblivious array access}, the access pattern on GPU is also protected.

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。

↑