Scalable23
From ACES
| Scalable23 | |
|---|---|
| entry | article |
| address | |
| annote | |
| author | Xinqiao Zhang and Mohammad Samragh and Siam Hussain and Ke Huang and Farinaz Koushanfar |
| booktitle | |
| chapter | |
| edition | |
| editor | |
| howpublished | |
| institution | |
| journal | |
| month | 7 |
| note | |
| number | |
| organization | |
| pages | |
| publisher | ACM Transactions on Embedded Computing Systems |
| school | |
| series | |
| title | Scalable Binary Neural Network applications in Oblivious Inference |
| type | |
| volume | |
| year | 2023 |
| doi | |
| issn | |
| isbn | |
| url | https://dl.acm.org/doi/pdf/10.1145/3607192 |
| Scalable23 | |
Binary neural network (BNN) delivers increased compute intensity and reduces memory/data requirements for computation. Scalable BNN enables inference in a limited time due to diferent constraints. This paper explores the application of Scalable BNN in oblivious inference, a service provided by a server to mistrusting clients. Using this service, a client can obtain the inference result on his/her data by a trained model held by the server without disclosing the data or learning the model parameters. Two contributions of this paper are: 1) we devise lightweight cryptographic protocols explicitly designed to exploit the unique characteristics of BNNs. 2) we present an advanced dynamic exploration of the runtime-accuracy tradeof of scalable BNNs in a single-shot training process. While previous works trained multiple BNNs with diferent computational complexities (which is cumbersome due to the slow convergence of BNNs), we train a single BNN that can perform inference under various computational budgets. Compared to CryptFlow2, the state-of-the-art technique in the oblivious inference of non-binary DNNs, our approach reaches 3× faster inference while keeping the same accuracy. Compared to XONN, the state-of-the-art technique in the oblivious inference of binary networks, we achieve 2× to 12× faster inference while obtaining higher accuracy.
farinaz@ucsd.edu
Electrical & Computer Engineering
University of California, San Diego
9500 Gilman Drive, MC 0407
Jacobs Hall, Room 6401
La Jolla, CA 92093-0407
University of California San Diego
9500 Gilman Dr, La Jolla, CA 92093