Difference between revisions of "Rouhani2018redcrypt"
From ACES
(Import from BibTeX) |
m (Import from BibTeX) |
||
| Line 3: | Line 3: | ||
|url=https://dl.acm.org/citation.cfm?id=3242899 | |url=https://dl.acm.org/citation.cfm?id=3242899 | ||
|abstract=<p><span style="color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">Artificial Intelligence (AI) is increasingly incorporated into the\ </span><i style="box-sizing: border-box; color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">cloud business</i><span style="color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">\ in order to improve the functionality (e.g., accuracy) of the service. The adoption of AI as a cloud service raises serious privacy concerns in applications where the\ </span><i style="box-sizing: border-box; color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">risk of data leakage</i><span style="color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">\ is not acceptable. Examples of such applications include scenarios where clients hold potentially sensitive private information such as medical records, financial data, and/or location. This article proposes ReDCrypt, the first\ </span><i style="box-sizing: border-box; color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">reconfigurable</i><span style="color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">\ hardware-accelerated framework that empowers privacy-preserving inference of deep learning models in cloud servers. ReDCrypt is well-suited for\ </span><i style="box-sizing: border-box; color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">streaming (a.k.a., real-time AI)</i><span style="color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">\ settings where clients need to dynamically analyze their data as it is collected over time without having to queue the samples to meet a certain batch size. Unlike prior work, ReDCrypt neither requires to change how AI models are trained nor relies on two non-colluding servers to perform. The privacy-preserving computation in ReDCrypt is executed using\ </span><i style="box-sizing: border-box; color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">Yao\’s Garbled Circuit</i><span style="color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">\ (GC) protocol. We break down the deep learning inference task into two phases: (i) privacy-insensitive (local) computation, and (ii) privacy-sensitive (interactive) computation. We devise a high-throughput and power-efficient implementation of GC protocol on FPGA for the privacy-sensitive phase. ReDCrypt\’s accompanying API provides support for seamless integration of ReDCrypt into any deep learning framework. Proof-of-concept evaluations for different DL applications demonstrate up to 57-fold higher throughput per core compared to the best prior solution with no drop in the accuracy.</span></p> | |abstract=<p><span style="color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">Artificial Intelligence (AI) is increasingly incorporated into the\ </span><i style="box-sizing: border-box; color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">cloud business</i><span style="color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">\ in order to improve the functionality (e.g., accuracy) of the service. The adoption of AI as a cloud service raises serious privacy concerns in applications where the\ </span><i style="box-sizing: border-box; color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">risk of data leakage</i><span style="color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">\ is not acceptable. Examples of such applications include scenarios where clients hold potentially sensitive private information such as medical records, financial data, and/or location. This article proposes ReDCrypt, the first\ </span><i style="box-sizing: border-box; color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">reconfigurable</i><span style="color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">\ hardware-accelerated framework that empowers privacy-preserving inference of deep learning models in cloud servers. ReDCrypt is well-suited for\ </span><i style="box-sizing: border-box; color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">streaming (a.k.a., real-time AI)</i><span style="color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">\ settings where clients need to dynamically analyze their data as it is collected over time without having to queue the samples to meet a certain batch size. Unlike prior work, ReDCrypt neither requires to change how AI models are trained nor relies on two non-colluding servers to perform. The privacy-preserving computation in ReDCrypt is executed using\ </span><i style="box-sizing: border-box; color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">Yao\’s Garbled Circuit</i><span style="color: rgb(0, 0, 0); font-family: Verdana, Arial, Helvetica, sans-serif;">\ (GC) protocol. We break down the deep learning inference task into two phases: (i) privacy-insensitive (local) computation, and (ii) privacy-sensitive (interactive) computation. We devise a high-throughput and power-efficient implementation of GC protocol on FPGA for the privacy-sensitive phase. ReDCrypt\’s accompanying API provides support for seamless integration of ReDCrypt into any deep learning framework. Proof-of-concept evaluations for different DL applications demonstrate up to 57-fold higher throughput per core compared to the best prior solution with no drop in the accuracy.</span></p> | ||
|month=12 | |||
|year=2018 | |||
|journal=ACM Transactions on Reconfigurable Technology and Systems (TRETS) - Special Issue on Deep learning on FPGAs | |journal=ACM Transactions on Reconfigurable Technology and Systems (TRETS) - Special Issue on Deep learning on FPGAs | ||
|title=ReDCrypt: Real-Time Privacy-Preserving Deep Learning Inference in Clouds Using FPGAs | |title=ReDCrypt: Real-Time Privacy-Preserving Deep Learning Inference in Clouds Using FPGAs | ||
|entry=article | |entry=article | ||
}} | }} | ||
Revision as of 03:45, 4 September 2021
| Rouhani2018redcrypt | |
|---|---|
| entry | article |
| address | |
| annote | |
| author | Rouhani, Bita D and Siam U. Hussain and Kristin Lauter and Farinaz Koushanfar |
| booktitle | |
| chapter | |
| edition | |
| editor | |
| howpublished | |
| institution | |
| journal | ACM Transactions on Reconfigurable Technology and Systems (TRETS) - Special Issue on Deep learning on FPGAs |
| month | 12 |
| note | |
| number | |
| organization | |
| pages | |
| publisher | |
| school | |
| series | |
| title | ReDCrypt: Real-Time Privacy-Preserving Deep Learning Inference in Clouds Using FPGAs |
| type | |
| volume | |
| year | 2018 |
| doi | |
| issn | |
| isbn | |
| url | https://dl.acm.org/citation.cfm?id=3242899 |
Email:
farinaz@ucsd.edu
farinaz@ucsd.edu
Address:
Electrical & Computer Engineering
University of California, San Diego
9500 Gilman Drive, MC 0407
Jacobs Hall, Room 6401
La Jolla, CA 92093-0407
Electrical & Computer Engineering
University of California, San Diego
9500 Gilman Drive, MC 0407
Jacobs Hall, Room 6401
La Jolla, CA 92093-0407
Lab Location: EBU1-2514
University of California San Diego
9500 Gilman Dr, La Jolla, CA 92093
University of California San Diego
9500 Gilman Dr, La Jolla, CA 92093