Homomorphic Encryption (HE) is a cryptographic tool that allows performing computation under encryption, which is used by many privacy-preserving machine learning solutions, for example, to perform secure classification. Modern deep learning applications yield good performance for example in image processing tasks benchmarks by including many skip connections. The latter appears to be very costly when attempting to execute model inference under HE. In this paper, we show that by replacing (mid-term) skip connections with (short-term) Dirac parameterization and (long-term) shared-source skip connection we were able to reduce the skip connections burden for HE-based solutions, achieving x1.3 computing power improvement for the same accuracy.

翻译：同态加密（HE）是一种允许在加密状态下执行计算的密码学工具，被许多隐私保护机器学习解决方案所采用，例如用于实现安全分类。现代深度学习应用通过包含大量跳跃连接，在图像处理等任务基准测试中取得了良好性能。然而，在尝试通过HE执行模型推理时，跳跃连接会带来极高的计算开销。本文证明，通过将（中期）跳跃连接替换为（短期）Dirac参数化与（长期）共享源跳跃连接，我们能够显著降低基于HE的方案中跳跃连接带来的计算负担，在保持相同精度的前提下实现1.3倍的算力提升。