GELU-Net: A Globally Encrypted, Locally Unencrypted Deep Neural Network for Privacy-Preserved Learning

Zhang, Qiao; Wang, Cong; Wu, Hongyi; Xin, Chunsheng; Phuong, Tran Viet Xuan

doi:10.24963/ijcai.2018/547

Cited by 60 publications

(58 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They consider collaborative training, i.e., the training is performed collaboratively between different participants or, individual training where the training is performed by a single participant, such as a client that wants to take advantage of a cloud to train its own model. In the latter case, some techniques ( [8], [52], and [31]) might allow a cloud model to be trained by successive participants. Noting that some collaborative learning techniques ( [20] and [23]), can also be applied for individual training.…”

Section: Classificationmentioning

confidence: 99%

“…A more recent solution based on encryption was described by Zhang et al [52]. A client, willing to contribute to the training of the model, sends its data encrypted using Paillier scheme to the server, which performs all possible neural network computations except of non-linear activation functions.…”

Section: B1) Server-based Individual Pp Model Learningmentioning

confidence: 99%

“…The result is then re-encrypted and sent again to the server. The authors of [52], on the basis of conducted experiments, considered the training as stable, and without accuracy loss. Moreover, the solution achieved a speed-up of 14 to 35 times in computation speed compared to CryptoNet.…”

Section: B1) Server-based Individual Pp Model Learningmentioning

confidence: 99%

See 2 more Smart Citations

A review of privacy-preserving techniques for deep learning

2020

View full text Add to dashboard Cite

Deep learning is one of the advanced approaches of machine learning, and has attracted a growing attention in the recent years. It is used nowadays in different domains and applications such as pattern recognition, medical prediction, and speech recognition. Differently from traditional learning algorithms, deep learning can overcome the dependency on hand-designed features. Deep learning experience is particularly improved by leveraging powerful infrastructures such as clouds and adopting collaborative learning for model training. However, this comes at the expense of privacy, especially when sensitive data are processed during the training and the prediction phases, as well as when training model is shared. In this paper, we provide a review of the existing privacy-preserving deep learning techniques, and propose a novel multilevel taxonomy, which categorizes the current state-of-the-art privacy-preserving deep learning techniques on the basis of privacy-preserving tasks at the top level, and key technological concepts at the base level. This survey further summarizes evaluation results of the reviewed solutions with respect to defined performance metrics. In addition, it derives a set of learned lessons from each privacy-preserving task. Finally, it highlights open research challenges and provides some recommendations as future research directions.

show abstract

Section: Classificationmentioning

confidence: 99%

Section: B1) Server-based Individual Pp Model Learningmentioning

confidence: 99%

See 1 more Smart Citation

A review of privacy-preserving techniques for deep learning

2020

View full text Add to dashboard Cite

show abstract

“…However, FHE is extremely costly in computation, thus unsuitable for large‐scale neural networks. To resolve this problem of existing FHE‐based approaches, Zhang et al proposed a new network named GELU‐Net, of which intrinsic strategy is to split each neuron into linear and nonlinear components and implement them separately on noncolluding parties.…”

Section: Related Workmentioning

confidence: 99%

Guaranteeing differential privacy for sequence predictions in bike sharing systems

Xie

Huang

et al. 2020

Concurrency and Computation

View full text Add to dashboard Cite

Summary Being part of public transportations, the bike sharing system plays an important role in solving the “last mile” problem. With more advantages than traditional bike sharing systems relying on fixed docks, dockless bike sharing systems are rapidly expanding, which generally operated under cloud‐based architectures. Emerging Things‐Edge‐Cloud (TEC) architectures are showing many advantages than cloud‐based architectures including greater potential in reducing risks of privacy leakage due to their decentralized computing manners. However, even in the TEC architecture, differential attacks, which aims to mine information of individuals in the aggregated data, still cause major threats. To solve this problem, this article proposes a novel encoding and decoding framework within the TEC architecture for time series predictions in bike sharing systems, with considerations of guaranteeing differential privacy. In particular, we first construct a dynamic autoencoder based on the Long Short Term Memory (LSTM) network, then the collected raw temporal data are encoded into a hidden state in the edge end. In the cloud end, the trained output weights are used to reconstruct the current time series and predict the next‐period time series according to received hidden states. This framework not only improves the computation efficiency of bike sharing system by leveraging computing power at the node end but also provides safer data publications that meet the differential privacy requirements. Experiments are conducted on real‐world datasets, results demonstrate the effectiveness of the proposed framework in providing data services with both high utility in sequence predictions and high safety level for users' privacy.

show abstract

“…For example, Cheon et al [5] proposed an ensemble method for logistic regreesion based on HE, which resulted in substantial improvement on the performance of logistic regression. Zhang et al [6] proposed GELU-Net, which was a novel privacy-preserving architecture where clients can collaboratively train a deep neural network model. Their experiments demonstrated the stability in training and time speed-up without accuracy loss.…”

Section: Introductionmentioning

confidence: 99%

Privacy-Preserved Approximate Classification Based on Homomorphic Encryption

Chen

Fan

2019

MCA

View full text Add to dashboard Cite

Privacy is a crucial issue for outsourcing computation, which means that clients utilize cloud infrastructure to perform online prediction without disclosing sensitive information. Homomorphic encryption (HE) is one of the promising cryptographic tools resolving privacy issue in this scenario. However, a bottleneck in application of HE is relatively high computational overhead. In this paper, we study the privacy-preserving classification problem. To this end, we propose a novel privacy-preserved approximate classification algorithm. It exploits a set of decision trees to reduce computational complexity during homomorphic evaluation computation formula, the time complexity of evaluating a polynomial is degraded from O n to O log n . As a result, for an MNIST dataset, the Micro- f 1 score of the proposed algorithm is 0 . 882 , compared with 0 . 912 of the standard method. For the Credit dataset, the algorithm achieves 0 . 601 compared with 0 . 613 of the method. These results show that our algorithm is feasible and practical in real world problems.

show abstract

GELU-Net: A Globally Encrypted, Locally Unencrypted Deep Neural Network for Privacy-Preserved Learning

Cited by 60 publications

References 12 publications

A review of privacy-preserving techniques for deep learning

A review of privacy-preserving techniques for deep learning

Guaranteeing differential privacy for sequence predictions in bike sharing systems

Privacy-Preserved Approximate Classification Based on Homomorphic Encryption

Contact Info

Product

Resources

About