nGraph-HE2：一个高性能的加密数据神经网络推理框架
nGraph-HE2: A High-Throughput Framework for Neural Network Inference on Encrypted Data
摘要：
在之前的工作中，Boemer等人介绍了nGraph-HE，这是英特尔nGraph深度学习（DL）编译器的扩展，使数据科学家能够使用流行的框架（如TensorFlow和PyTorch）部署模型，只需最少的代码更改。然而，这类支持的模型仅限于具有多项式激活的相对浅的网络。在这里，我们介绍了nGraph-HE 2，它扩展了nGraph-HE，可以使用其原生激活函数和数字字段（通常是实数）对标准的预训练模型进行隐私保护推理。所提出的框架利用CKKS方案，其对实数的支持对数据科学是友好的，以及使用两方方法来计算激活函数的客户端辅助模型。我们首先提出了CKKS特定的优化，使标量编码的3x-88 x的运行时加速，并通过新的使用CKKS明文包装成复数的吞吐量增加一倍。其次，我们优化了密文-明文加法和乘法，产生2.6x-4.2x的运行时加速。第三，我们利用两个图级优化：延迟重新缩放和深度感知编码，这使我们能够显着提高性能。这些优化共同实现了CryptoNets网络上1，998张图像/秒的最先进吞吐量。使用客户端辅助模型，我们还对迄今为止最大的网络进行了同态评估（据我们所知），即ImageNet数据集上的预训练MobileNetV 2模型，具有60.4 \percent /82.7 \percent \ top-1/top-5的准确性和381 ms/图像的摊销运行时间。
英文摘要：
In previous work, Boemer et al. introduced nGraph-HE, an extension to the Intel nGraph deep learning (DL) compiler, that enables data scientists to deploy models with popular frameworks such as TensorFlow and PyTorch with minimal code changes. However, the class of supported models was limited to relatively shallow networks with polynomial activations. Here, we introduce nGraph-HE2, which extends nGraph-HE to enable privacy-preserving inference on standard, pre-trained models using their native activation functions and number fields (typically real numbers). The proposed framework leverages the CKKS scheme, whose support for real numbers is friendly to data science, and a client-aided model using a two-party approach to compute activation functions. We first present CKKS-specific optimizations, enabling a 3x-88x runtime speedup for scalar encoding, and doubling the throughput through a novel use of CKKS plaintext packing into complex numbers. Second, we optimize ciphertext-plaintext addition and multiplication, yielding 2.6x-4.2x runtime speedup. Third, we exploit two graph-level optimizations: lazy rescaling and depth-aware encoding, which allow us to significantly improve performance. Together, these optimizations enable state-of-the-art throughput of 1,998 images/s on the CryptoNets network. Using the client-aided model, we also present homomorphic evaluation of (to our knowledge) the largest network to date, namely, pre-trained MobileNetV2 models on the ImageNet dataset, with 60.4\percent/82.7\percent\ top-1/top-5 accuracy and an amortized runtime of 381 ms/image.
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。
