EI-DDLGN：TFHE下基于深度可微逻辑门网络的高效加密推理
EI-DDLGN: Efficient Encrypted Inference with Deep Differentiable Logic Gate Networks under TFHE
浏览论文内容
中文总结 AI 辅助
提出EI-DDLGN，首次深入研究TFHE下DDLGN加密推理，通过MFW-PBS旁路减少PBS操作，在MNIST等数据集上实现比算术TFHE推理更优的精度-延迟权衡，延迟降低13.4倍。
中文摘要 AI 辅助
通过环面全同态加密（TFHE）进行的隐私保护推理，为外包深度学习应用中的敏感数据提供了强有力的保护。然而，大多数兼容TFHE的神经网络框架仍基于算术神经架构，导致由于可编程自举（PBS）、累加器增长和电路位宽敏感性而产生高推理延迟。在本工作中，我们研究了深度可微逻辑门网络（DDLGNs）作为TFHE下加密推理的布尔原生替代方案。由于DDLGNs直接学习布尔计算并离散化为固定的逻辑门网络，其推理过程自然与TFHE的布尔执行模型对齐，并避免了隐藏层中的算术累加。我们提出了EI-DDLGN，这是对基于TFHE的DDLGN推理的首次深入研究，并刻画了加密执行成本如何依赖于模型大小、学习到的布尔函数分布和传播的线状态。我们还引入了模型固定线PBS旁路（MFW-PBS旁路），这是一种保持语义的执行策略，在不修改学习到的网络拓扑的情况下消除不必要的PBS操作。在MNIST、FashionMNIST和UCI Phishing上跨越72种深度-宽度配置的评估表明，DDLGNs构成了算术TFHE推理的高效替代方案，实现了显著改善的精度-延迟权衡。值得注意的是，在MNIST上，EI-DDLGN-Small匹配了QAT-FCNN-4的精度，同时将加密推理延迟降低了13.4倍。我们的实现可在以下网址获取：https://this https URL
英文摘要
Privacy-preserving inference via Torus Fully Homomorphic Encryption (TFHE) provides strong protection for sensitive data in outsourced deep learning applications. However, most TFHE-compatible neural network frameworks remain based on arithmetic neural architectures, resulting in high inference latency due to programmable bootstrapping (PBS), accumulator growth, and circuit bit-width sensitivity. In this work, we investigate Deep Differentiable Logic Gate Networks (DDLGNs) as a Boolean-native alternative for encrypted inference under TFHE. Because DDLGNs learn Boolean computations directly and discretize into fixed logic gate networks, their inference procedure is naturally aligned with TFHE's Boolean execution model and avoids arithmetic accumulation in hidden layers. We present EI-DDLGN, the first in-depth study of TFHE-based DDLGN inference, and characterize how encrypted execution cost depends on model size, learned Boolean-function distribution, and propagated wire status. We also introduce Model-Fixed-Wire PBS Bypass (MFW-PBS Bypass), a semantics-preserving execution strategy that eliminates unnecessary PBS operations without modifying the learned network topology. Evaluations across 72 depth-width configurations on MNIST, FashionMNIST, and UCI Phishing show that DDLGNs constitute an efficient alternative to arithmetic TFHE inference, achieving substantially improved accuracy-latency trade-offs. Notably, on MNIST, EI-DDLGN-Small matches the accuracy of QAT-FCNN-4 while reducing encrypted inference latency by 13.4x. Our implementation is available at https://github.com/Carleton-SCI/EI-DDLGN
发表机构
- Carleton University（卡尔顿大学）
机构由 AI 辅助整理，请以论文原文为准。
