SynCirc: Efficient Synthesis of Depth-Optimized Circuits for Secure Computation

Patra, Arpita; Schneider, Thomas; Suresh, Ananda Theertha; Yalame, Hossein

doi:10.1109/host49136.2021.9702293

Cited by 33 publications

(78 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this study, we adopt a new sharing semantics of the work of Arpita et al 25 to significantly reduce rounds and communication overhead and gradually extend it to all computing operations.…”

Section: Preliminariesmentioning

confidence: 99%

“…The difference between two protocols is that

Mul{t}^{* }

does not require reconstruction. Naturally, the protocol

Mult

can extend to 3‐input or even

n

‐input multiplication, 25 and the protocol

preMult

will match protocol

Mult

correspondingly.…”

Section: Preliminariesmentioning

confidence: 99%

“…The term

{[{\delta }_{x}]}_{0}{[{\delta }_{y}]}_{1}

and

{[{\delta }_{x}]}_{1}{[{\delta }_{y}]}_{0}

are computed as follows. Based on the works of OT, 25,39,42–44 the sender inputs a function

f(\cdot )

to obtain

({s}_{0},{s}_{1}=f({s}_{0}))

. We use

{\,\text{C\unicode{x02010}OT}\,}_{\ell }^{\ell }

to represent

\ell

parallel instances of 1‐out‐of‐2 correlated OTs on

\ell

‐bit strings.…”

Section: Preliminariesmentioning

confidence: 99%

“…In the research of using a single secret sharing technique to solve PPML issues, the works 19–24 use the traditional additive secret sharing technique, 5 which makes the multiplication operation need two reconstructions, however, the large number of multiplication operations in the training and inference process of CNN certainly put great pressure on the communication ability. In addition, for the nonlinear activation functions such as softmax, sigmoid, and tanh, the works 21,25 are to find MPC‐friendly expressions to fit them, such as piecewise function, polynomial fitting, and so on. However, these solutions are difficult to break through the accuracy bottleneck.…”

Section: Introductionmentioning

confidence: 99%

“…In this study, we propose a framework for privacy‐preserving CNN (PPCNN) based on the secret sharing technique introduced in ABY2.0 25 to protect the privacy of user data and model parameters during CNN training and inference. Before the CNN model training starts, the data owners encrypt data with the secret sharing technique and send two different data shares to two noncolluding cloud servers, respectively.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

PPCNN: An efficient privacy‐preserving CNN training and inference framework

Zhao

Wang

2022

Int J of Intelligent Sys

View full text Add to dashboard Cite

Convolutional neural network (CNN) is one of the representative models of deep learning, commonly used to analyze visual images. CNN model is more accurate when trained on large amounts of data from multiple sources, and the huge training cost makes the model much more valuable. However, data from various sources is often privacy‐sensitive. Therefore, the privacy of these data should be protected during CNN model training and inference. In this paper, we propose an efficient and secure two‐party computation (2PC) framework PPCNN for privacy‐preserving CNN training and inference. Specifically, we use a new secret sharing technique introduced in ABY2.0 to securely compute various computational tasks involved in the CNN training and inference processes. This secret sharing technique can significantly reduce the communication overhead. Meanwhile, we assign these computationally intensive tasks to cloud servers to reduce the computational burden on local devices. We demonstrate the security of these protocols in the semihonest model. In addition, we use the MP‐SPDZ library to simulate our PPCNN framework, and the experiments prove its high efficiency and accuracy.

show abstract

Section: Preliminariesmentioning

confidence: 99%

“…The difference between two protocols is that

Mul{t}^{* }

does not require reconstruction. Naturally, the protocol

Mult

can extend to 3‐input or even

n

‐input multiplication, 25 and the protocol

preMult

will match protocol

Mult

correspondingly.…”

Section: Preliminariesmentioning

confidence: 99%

“…The term

{[{\delta }_{x}]}_{0}{[{\delta }_{y}]}_{1}

and

{[{\delta }_{x}]}_{1}{[{\delta }_{y}]}_{0}

are computed as follows. Based on the works of OT, 25,39,42–44 the sender inputs a function

f(\cdot )

to obtain

({s}_{0},{s}_{1}=f({s}_{0}))

. We use

{\,\text{C\unicode{x02010}OT}\,}_{\ell }^{\ell }

to represent

\ell

parallel instances of 1‐out‐of‐2 correlated OTs on

\ell

‐bit strings.…”

Section: Preliminariesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

PPCNN: An efficient privacy‐preserving CNN training and inference framework

Zhao

Wang

2022

Int J of Intelligent Sys

View full text Add to dashboard Cite

show abstract

Breaking the Size Barrier: Universal Circuits Meet Lookup Tables

Disser,

Günther,

Schneider

et al. 2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

PPML-Omics: a Privacy-Preserving federated Machine Learning method protects patients’ privacy in omic data

Zhou

Chen

et al. 2022

Preprint

View full text Add to dashboard Cite

Modern machine learning models towards various tasks with omic data analysis give rise to threats of privacy leakage of patients involved in those datasets. Despite the advances in different privacy technologies, existing methods tend to introduce too much noise, which hampers model accuracy and usefulness. Here, we built a secure and privacy-preserving machine learning (PPML) system by combining federated learning (FL), differential privacy (DP) and shuffling mechanism. We applied this system to analyze data from three sequencing technologies, and addressed the privacy concern in three major tasks of omic data, namely cancer classification with bulk RNA-seq, clustering with single-cell RNA-seq, and the integration of spatial gene expression and tumour morphology with spatial transcriptomics, under three representative deep learning models. We also examined privacy breaches in depth through privacy attack experiments and demonstrated that our PPML-Omics system could protect patients' privacy. In each of these applications, PPML-Omics was able to outperform state-of-the-art systems under the same level of privacy guarantee, demonstrating the versatility of the system in simultaneously balancing the privacy-preserving capability and utility in omic data analysis. Furthermore, we gave the theoretical proof of the privacy-preserving capability of PPML-Omics, suggesting the first mathematically guaranteed model with robust and generalizable empirical performance.

show abstract

SynCirc: Efficient Synthesis of Depth-Optimized Circuits for Secure Computation

Cited by 33 publications

References 49 publications

PPCNN: An efficient privacy‐preserving CNN training and inference framework

PPCNN: An efficient privacy‐preserving CNN training and inference framework

Breaking the Size Barrier: Universal Circuits Meet Lookup Tables

PPML-Omics: a Privacy-Preserving federated Machine Learning method protects patients’ privacy in omic data

Contact Info

Product

Resources

About