通过联合优化网络量化与安全协议，显著降低私有推理通信开销。
PrivQuant: Communication-Efficient Private Inference with Quantized Network/Protocol Co-Optimization
- 协同优化量化算法与2PC协议，减少通信量。
- 通信量降低11倍，延迟降低8.7倍，精度保持高水准。
- 适合对隐私和效率要求高的医疗、金融等场景使用。
基于安全两方计算（2PC）的私有深度神经网络（DNN）推理可同时保护服务器和客户端隐私。然而，现有2PC框架因通信开销大导致推理延迟高。由于线性与非线性层的通信量随权重和激活值的位宽降低而减少，本文提出PrivQuant框架，联合优化基于2PC的量化推理协议与网络量化算法，实现通信高效的私有推理。PrivQuant针对通信密集型量化算子设计了面向DNN架构的2PC协议优化，并进行图级算子融合以进一步降低通信开销。此外，还提出了通信感知的混合精度量化算法，在保持高精度的同时提升推理效率。实验表明，相比SiRNN、COINN和CoPriv，PrivQuant分别实现11×、2.5×、2.8×的通信量减少，对应延迟降低8.7×、1.8×、2.4×。
原文摘要 · Abstract (English)
Private deep neural network (DNN) inference based on secure two-party computation (2PC) enables secure privacy protection for both the server and the client. However, existing secure 2PC frameworks suffer from a high inference latency due to enormous communication. As the communication of both linear and non-linear DNN layers reduces with the bit widths of weight and activation, in this paper, we propose PrivQuant, a framework that jointly optimizes the 2PC-based quantized inference protocols and the network quantization algorithm, enabling communication-efficient private inference. PrivQuant proposes DNN architecture-aware optimizations for the 2PC protocols for communication-intensive quantized operators and conducts graph-level operator fusion for communication reduction. Moreover, PrivQuant also develops a communication-aware mixed precision quantization algorithm to improve inference efficiency while maintaining high accuracy. The network/protocol co-optimization enables PrivQuant to outperform prior-art 2PC frameworks. With extensive experiments, we demonstrate PrivQuant reduces communication by $11\times, 2.5\times \mathrm{and}~ 2.8\times$, which results in $8.7\times, 1.8\times ~ \mathrm{and}~ 2.4\times$ latency reduction compared with SiRNN, COINN, and CoPriv, respectively.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。