CryptoNite:揭示大规模端到端私有推理的陷阱
CryptoNite: Revealing the Pitfalls of End-to-End Private Inference at Scale
摘要:
将深度学习推理作为服务提供的隐私问题强调了对使用加密方法保护用户数据和服务提供商模型的私有推理(PI)协议的需求。最近提出的PI协议通过将计算量大的同态加密(HE)部分移动到离线/预计算阶段来实现PI延迟的显著减少。再加上最近针对PI定制网络的优化,这些协议已经达到了非常接近实用的性能水平。在本文中,我们进行了严格的端到端的PI协议和优化技术的表征,发现目前的理解PI性能过于乐观。具体而言,我们发现,在用户/客户端设备上的乱码电路(GC),PI中使用的密钥加密协议的离线存储成本非常高,并迫使许多昂贵的离线HE计算到在线阶段,导致10-1000 $\times$增加PI延迟。我们提出了一个修改后的PI协议,显着降低了客户端的存储成本,在线延迟的小增加。端到端评估,修改后的协议通过将TinyImageNet上ResNet 18的平均PI延迟减少$4\times$而优于当前协议。最后,我们根据研究结果讨论了最近提出的几种PI优化,并注意到从端到端的角度评估时,许多PI优化实际上增加了PI延迟。
英文摘要:
The privacy concerns of providing deep learning inference as a service have underscored the need for private inference (PI) protocols that protect users' data and the service provider's model using cryptographic methods. Recently proposed PI protocols have achieved significant reductions in PI latency by moving the computationally heavy homomorphic encryption (HE) parts to an offline/pre-compute phase. Paired with recent optimizations that tailor networks for PI, these protocols have achieved performance levels that are tantalizingly close to being practical. In this paper, we conduct a rigorous end-to-end characterization of PI protocols and optimization techniques and find that the current understanding of PI performance is overly optimistic. Specifically, we find that offline storage costs of garbled circuits (GC), a key cryptographic protocol used in PI, on user/client devices are prohibitively high and force much of the expensive offline HE computation to the online phase, resulting in a 10-1000$\times$ increase to PI latency. We propose a modified PI protocol that significantly reduces client-side storage costs for a small increase in online latency. Evaluated end-to-end, the modified protocol outperforms current protocols by reducing the mean PI latency by $4\times$ for ResNet18 on TinyImageNet. We conclude with a discussion of several recently proposed PI optimizations in light of the findings and note many actually increase PI latency when evaluated from an end-to-end perspective.
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。