Secure multiple linear regression based on homomorphic encryption

By Rob Hall, Stephen E. Fienberg, Yuval Nardi. 2011

143 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Federated Machine Learning (2019, background): “ons during training. Homomorphic Encryption. Homomorphic Encryption [53] is also adopted to protect user data privacy through parameter exchange under the encryption mechanism during machine learning [24, 26, 48]. Unlike differential privacy protection, the data and the model itself are not transmit”
Privacy-Preserving Distributed Linear Regression on High-Dimensional Data (2017, methodology): “As in previous work on privacy-preserving ridge regression [35, 59], we assume that the regularization parameter λ is public to all the parties, and hence the implementation of a secure data analysis pipeline including hyper-parameter tuning (e.”
Privacy-Preserving Ridge Regression on Hundreds of Millions of Records (2013, methodology): “20 25 30 35 40 0 20 40 60 80 100 120 140 Total bits Execution and garbling time (s) 14 features 12 features 10 features 8 features 6 features 4 features 2 features Figure 7: Time for execution of garbled circuit depending on number of bits and number of features E. Comparison to State of the Art Hal”
Privacy-preserving matrix factorization (2013, background): “Secret-sharing schemes have been proposed for a variety of linear algebra operations, such as solving a linear system [51], linear regression [29, 30, 21], and auctions [10].”
Fast, Privacy Preserving Linear Regression over Distributed Datasets based on Pre-Distributed Data (2015, methodology): “We improve the running time of previous algorithms for secure linear regression from days [19] to seconds in the online phase.”
Privacy-Preserving Statistical Analysis by Exact Logistic Regression (2015, methodology): “proposed a secure protocol to learn the parameters of a logistic regression model [28], [29], [2].”
Faster CryptoNets: Leveraging Sparsity for Real-World Encrypted Inference (2018, methodology): “While HE has been explored for machine learning applications, many works focus on simpler models such as linear [34], logistic [18] and ridge regression [28].”
Secure Linear Regression on Vertically Partitioned Datasets (2016, methodology): “Our fixed-point encoding strategy follows closely previous works on secure linear regression [9, 19, 35].”
Encrypted Accelerated Least Squares Regression (2017, background): “Hall et al. (2011) propose protocols for regression analysis which involve substantial communication and intermediate decryption between multiple parties: it takes two days to complete a problem of size N = 51, 016 observations and P = 23 predictors.”
Privacy Preserving Machine Learning with Homomorphic Encryption and Federated Learning (2021, background): “Based on privacy protection, machine learning with multi-party privacy protection could help users of all parties to jointly learn with each other’s data, on the premise of ensuring the security of their own data [1–3].”
A Survey on Federated Learning Systems: Vision, Hype and Reality for Data Privacy and Protection (2019, methodology): “[179]. Here we introduce two major approaches that are adopted in current FLSs for data protection: cryptographic methods and differential privacy. Cryptographic methods such as homomorphic encryption[15, 77, 27, 33, 74, 145, 146, 204, 206, 112], and secure multi-party computation (SMC) [157, 36, 23”