CBNN: 3-Party Secure Framework for Customized Binary Neural Networks Inference
This work addresses privacy-preserving machine learning for scenarios requiring efficient and secure BNN inference, representing an incremental improvement with specific optimizations.
The paper tackles the challenges of communication and accuracy in secure BNN inference by introducing CBNN, a three-party secure computation framework that transforms standard BNNs into MPC-friendly customized versions using knowledge distillation and separable convolutions, achieving impressive performance with experimental results showing maintained utility after customization and security measures.
Binarized Neural Networks (BNN) offer efficient implementations for machine learning tasks and facilitate Privacy-Preserving Machine Learning (PPML) by simplifying operations with binary values. Nevertheless, challenges persist in terms of communication and accuracy in their application scenarios. In this work, we introduce CBNN, a three-party secure computation framework tailored for efficient BNN inference. Leveraging knowledge distillation and separable convolutions, CBNN transforms standard BNNs into MPC-friendly customized BNNs, maintaining high utility. It performs secure inference using optimized protocols for basic operations. Specifically, CBNN enhances linear operations with replicated secret sharing and MPC-friendly convolutions, while introducing a novel secure activation function to optimize non-linear operations. We demonstrate the effectiveness of CBNN by transforming and securely implementing several typical BNN models. Experimental results indicate that CBNN maintains impressive performance even after customized binarization and security measures