CBNN: 3-Party Secure Framework for Customized Binary Neural Networks Inference
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912165456248832 |
|---|---|
| author | Dong, Benchang Chen, Zhili Chen, Xin Wei, Shiwen Fu, Jie Li, Huifa |
| author_facet | Dong, Benchang Chen, Zhili Chen, Xin Wei, Shiwen Fu, Jie Li, Huifa |
| contents | 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 |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_16449 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | CBNN: 3-Party Secure Framework for Customized Binary Neural Networks Inference Dong, Benchang Chen, Zhili Chen, Xin Wei, Shiwen Fu, Jie Li, Huifa Machine Learning Cryptography and Security 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 |
| title | CBNN: 3-Party Secure Framework for Customized Binary Neural Networks Inference |
| topic | Machine Learning Cryptography and Security |
| url | https://arxiv.org/abs/2412.16449 |