Low Latency Privacy Preserving Inference

Alon Brutzkus, Ran Gilad-Bachrach, Oren Elisha
Proceedings of the 36th International Conference on Machine Learning, PMLR 97:812-821, 2019.

Abstract

When applying machine learning to sensitive data, one has to find a balance between accuracy, information security, and computational-complexity. Recent studies combined Homomorphic Encryption with neural networks to make inferences while protecting against information leakage. However, these methods are limited by the width and depth of neural networks that can be used (and hence the accuracy) and exhibit high latency even for relatively simple networks. In this study we provide two solutions that address these limitations. In the first solution, we present more than 10\times improvement in latency and enable inference on wider networks compared to prior attempts with the same level of security. The improved performance is achieved by novel methods to represent the data during the computation. In the second solution, we apply the method of transfer learning to provide private inference services using deep networks with latency of \sim0.16 seconds. We demonstrate the efficacy of our methods on several computer vision tasks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v97-brutzkus19a, title = {Low Latency Privacy Preserving Inference}, author = {Brutzkus, Alon and Gilad-Bachrach, Ran and Elisha, Oren}, booktitle = {Proceedings of the 36th International Conference on Machine Learning}, pages = {812--821}, year = {2019}, editor = {Chaudhuri, Kamalika and Salakhutdinov, Ruslan}, volume = {97}, series = {Proceedings of Machine Learning Research}, month = {09--15 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v97/brutzkus19a/brutzkus19a.pdf}, url = {https://proceedings.mlr.press/v97/brutzkus19a.html}, abstract = {When applying machine learning to sensitive data, one has to find a balance between accuracy, information security, and computational-complexity. Recent studies combined Homomorphic Encryption with neural networks to make inferences while protecting against information leakage. However, these methods are limited by the width and depth of neural networks that can be used (and hence the accuracy) and exhibit high latency even for relatively simple networks. In this study we provide two solutions that address these limitations. In the first solution, we present more than 10\times improvement in latency and enable inference on wider networks compared to prior attempts with the same level of security. The improved performance is achieved by novel methods to represent the data during the computation. In the second solution, we apply the method of transfer learning to provide private inference services using deep networks with latency of \sim0.16 seconds. We demonstrate the efficacy of our methods on several computer vision tasks.} }
Endnote
%0 Conference Paper %T Low Latency Privacy Preserving Inference %A Alon Brutzkus %A Ran Gilad-Bachrach %A Oren Elisha %B Proceedings of the 36th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2019 %E Kamalika Chaudhuri %E Ruslan Salakhutdinov %F pmlr-v97-brutzkus19a %I PMLR %P 812--821 %U https://proceedings.mlr.press/v97/brutzkus19a.html %V 97 %X When applying machine learning to sensitive data, one has to find a balance between accuracy, information security, and computational-complexity. Recent studies combined Homomorphic Encryption with neural networks to make inferences while protecting against information leakage. However, these methods are limited by the width and depth of neural networks that can be used (and hence the accuracy) and exhibit high latency even for relatively simple networks. In this study we provide two solutions that address these limitations. In the first solution, we present more than 10\times improvement in latency and enable inference on wider networks compared to prior attempts with the same level of security. The improved performance is achieved by novel methods to represent the data during the computation. In the second solution, we apply the method of transfer learning to provide private inference services using deep networks with latency of \sim0.16 seconds. We demonstrate the efficacy of our methods on several computer vision tasks.
APA
Brutzkus, A., Gilad-Bachrach, R. & Elisha, O.. (2019). Low Latency Privacy Preserving Inference. Proceedings of the 36th International Conference on Machine Learning, in Proceedings of Machine Learning Research 97:812-821 Available from https://proceedings.mlr.press/v97/brutzkus19a.html.

Related Material