Efficient Representations for Privacy-Preserving Inference

Han Xuanyuan; Francisco Vargas; Stephen Cummins

Back to publications

Efficient Representations for Privacy-Preserving Inference

Han Xuanyuan, Francisco Vargas, Stephen Cummins

, 2021.

Abstract

Deep neural networks have a wide range of applications across multiple domains such as computer vision and medicine. In many cases, the input of a model at inference time can consist of sensitive user data, which raises questions concerning the levels of privacy and trust guaranteed by such services. Much existing work has leveraged homomorphic encryption (HE) schemes that enable computation on encrypted data to achieve private inference for multi-layer perceptrons and CNNs. An early work along this direction was CryptoNets, which takes 250 seconds for one MNIST inference. The main limitation of such approaches is that of compute, which is due to the costly nature of the NTT (number theoretic transform)operations that constitute HE operations. Others have proposed the use of model pruning and efficient data representations to reduce the number of HE operations required. In this paper, we focus on improving upon existing work by proposing changes to the representations of intermediate tensors during CNN inference. We construct and evaluate private CNNs on the MNIST and CIFAR-10 datasets, and achieve over a two-fold reduction in the number of operations used for inferences of the CryptoNets architecture.

Links

Cite this Paper

BibTeX


@Misc{publications/efficient-representations-for-privacy-preserving-inference,
  title = 	 {Efficient Representations for Privacy-Preserving Inference},
  author = 	 {Xuanyuan, Han and Vargas, Francisco and Cummins, Stephen},
  year = 	 {2021},
  url = 	 {/publications/efficient-representations-for-privacy-preserving-inference.html},
  abstract = 	 {Deep neural networks have a wide range of applications across multiple domains such as computer vision and medicine. In many cases, the input of a model at inference time can consist of sensitive user data, which raises questions concerning the levels of privacy and trust guaranteed by such services. Much existing work has leveraged homomorphic encryption (HE) schemes that enable computation on encrypted data to achieve private inference for multi-layer perceptrons and CNNs. An early work along this direction was CryptoNets, which takes 250 seconds for one MNIST inference. The main limitation of such approaches is that of compute, which is due to the costly nature of the NTT (number theoretic transform)operations that constitute HE operations. Others have proposed the use of model pruning and efficient data representations to reduce the number of HE operations required. In this paper, we focus on improving upon existing work by proposing changes to the representations of intermediate tensors during CNN inference. We construct and evaluate private CNNs on the MNIST and CIFAR-10 datasets, and achieve over a two-fold reduction in the number of operations used for inferences of the CryptoNets architecture.}
}

Endnote

%0 Generic
%T Efficient Representations for Privacy-Preserving Inference
%A Han Xuanyuan
%A Francisco Vargas
%A Stephen Cummins
%D 2021	
%F publications/efficient-representations-for-privacy-preserving-inference
%U /publications/efficient-representations-for-privacy-preserving-inference.html
%X Deep neural networks have a wide range of applications across multiple domains such as computer vision and medicine. In many cases, the input of a model at inference time can consist of sensitive user data, which raises questions concerning the levels of privacy and trust guaranteed by such services. Much existing work has leveraged homomorphic encryption (HE) schemes that enable computation on encrypted data to achieve private inference for multi-layer perceptrons and CNNs. An early work along this direction was CryptoNets, which takes 250 seconds for one MNIST inference. The main limitation of such approaches is that of compute, which is due to the costly nature of the NTT (number theoretic transform)operations that constitute HE operations. Others have proposed the use of model pruning and efficient data representations to reduce the number of HE operations required. In this paper, we focus on improving upon existing work by proposing changes to the representations of intermediate tensors during CNN inference. We construct and evaluate private CNNs on the MNIST and CIFAR-10 datasets, and achieve over a two-fold reduction in the number of operations used for inferences of the CryptoNets architecture.

RIS


TY  - GEN
TI  - Efficient Representations for Privacy-Preserving Inference
AU  - Han Xuanyuan
AU  - Francisco Vargas
AU  - Stephen Cummins
DA  - 2021/10/15	
ID  - publications/efficient-representations-for-privacy-preserving-inference
UR  - /publications/efficient-representations-for-privacy-preserving-inference.html
AB  - Deep neural networks have a wide range of applications across multiple domains such as computer vision and medicine. In many cases, the input of a model at inference time can consist of sensitive user data, which raises questions concerning the levels of privacy and trust guaranteed by such services. Much existing work has leveraged homomorphic encryption (HE) schemes that enable computation on encrypted data to achieve private inference for multi-layer perceptrons and CNNs. An early work along this direction was CryptoNets, which takes 250 seconds for one MNIST inference. The main limitation of such approaches is that of compute, which is due to the costly nature of the NTT (number theoretic transform)operations that constitute HE operations. Others have proposed the use of model pruning and efficient data representations to reduce the number of HE operations required. In this paper, we focus on improving upon existing work by proposing changes to the representations of intermediate tensors during CNN inference. We construct and evaluate private CNNs on the MNIST and CIFAR-10 datasets, and achieve over a two-fold reduction in the number of operations used for inferences of the CryptoNets architecture.
ER  -

APA


Xuanyuan, H., Vargas, F. & Cummins, S.. (2021). Efficient Representations for Privacy-Preserving Inference.  Available from /publications/efficient-representations-for-privacy-preserving-inference.html.