Round-Efficient Secure Inference Based on Masked Secret Sharing for Quantized Neural Network
Existing secure multiparty computation protocol from secret sharing is usually under this assumption of the fast network, which limits the practicality of the scheme on the low bandwidth and high latency network. A proven method is to reduce the communication rounds of the protocol as much as possib...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Entropy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1099-4300/25/2/389 |
| _version_ | 1797621122316894208 |
|---|---|
| author | Weiming Wei Chunming Tang Yucheng Chen |
| author_facet | Weiming Wei Chunming Tang Yucheng Chen |
| author_sort | Weiming Wei |
| collection | DOAJ |
| description | Existing secure multiparty computation protocol from secret sharing is usually under this assumption of the fast network, which limits the practicality of the scheme on the low bandwidth and high latency network. A proven method is to reduce the communication rounds of the protocol as much as possible or construct a constant-round protocol. In this work, we provide a series of constant-round secure protocols for quantized neural network (QNN) inference. This is given by masked secret sharing (MSS) in the three-party honest-majority setting. Our experiment shows that our protocol is practical and suitable for low-bandwidth and high-latency networks. To the best of our knowledge, this work is the first one where the QNN inference based on masked secret sharing is implemented. |
| first_indexed | 2024-03-11T08:52:16Z |
| format | Article |
| id | doaj.art-db861b3830f341d69d0d901a8e3d1baf |
| institution | Directory Open Access Journal |
| issn | 1099-4300 |
| language | English |
| last_indexed | 2024-03-11T08:52:16Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Entropy |
| spelling | doaj.art-db861b3830f341d69d0d901a8e3d1baf2023-11-16T20:24:49ZengMDPI AGEntropy1099-43002023-02-0125238910.3390/e25020389Round-Efficient Secure Inference Based on Masked Secret Sharing for Quantized Neural NetworkWeiming Wei0Chunming Tang1Yucheng Chen2School of Mathematics and Information Science, Guangzhou University, Guangzhou 510006, ChinaSchool of Mathematics and Information Science, Guangzhou University, Guangzhou 510006, ChinaSchool of Mathematics, Jiaying University, Meizhou 514015, ChinaExisting secure multiparty computation protocol from secret sharing is usually under this assumption of the fast network, which limits the practicality of the scheme on the low bandwidth and high latency network. A proven method is to reduce the communication rounds of the protocol as much as possible or construct a constant-round protocol. In this work, we provide a series of constant-round secure protocols for quantized neural network (QNN) inference. This is given by masked secret sharing (MSS) in the three-party honest-majority setting. Our experiment shows that our protocol is practical and suitable for low-bandwidth and high-latency networks. To the best of our knowledge, this work is the first one where the QNN inference based on masked secret sharing is implemented.https://www.mdpi.com/1099-4300/25/2/389secure inferencequantized neural networkmasked secret sharing |
| spellingShingle | Weiming Wei Chunming Tang Yucheng Chen Round-Efficient Secure Inference Based on Masked Secret Sharing for Quantized Neural Network Entropy secure inference quantized neural network masked secret sharing |
| title | Round-Efficient Secure Inference Based on Masked Secret Sharing for Quantized Neural Network |
| title_full | Round-Efficient Secure Inference Based on Masked Secret Sharing for Quantized Neural Network |
| title_fullStr | Round-Efficient Secure Inference Based on Masked Secret Sharing for Quantized Neural Network |
| title_full_unstemmed | Round-Efficient Secure Inference Based on Masked Secret Sharing for Quantized Neural Network |
| title_short | Round-Efficient Secure Inference Based on Masked Secret Sharing for Quantized Neural Network |
| title_sort | round efficient secure inference based on masked secret sharing for quantized neural network |
| topic | secure inference quantized neural network masked secret sharing |
| url | https://www.mdpi.com/1099-4300/25/2/389 |
| work_keys_str_mv | AT weimingwei roundefficientsecureinferencebasedonmaskedsecretsharingforquantizedneuralnetwork AT chunmingtang roundefficientsecureinferencebasedonmaskedsecretsharingforquantizedneuralnetwork AT yuchengchen roundefficientsecureinferencebasedonmaskedsecretsharingforquantizedneuralnetwork |