Distributed Learning without Distress: Privacy-Preserving Empirical Risk Minimization

By Bargav Jayaraman, Lingxiao Wang, David E. Evans, Quanquan Gu. 2018

174 citations recorded by the source corpus.

Subfields: Federated & Distributed Learning, Privacy & Security

How researchers see this paper

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

In Federated & Distributed Learning

In Privacy & Security

Citations in the corpus

Privacy Amplification by Decentralization (2020, background): “A popular approach is to resort to cryptographic primitives to securely aggregate user contributions (Dwork et al., 2006a; Shi et al., 2011; Bonawitz et al., 2017; Chan et al., 2012b; Jayaraman et al., 2018; Bell et al., 2020; Sabater et al., 2020) or to securely shuffle the set of user messages so ”
Differentially Private Learning Needs Better Features (or Much More Data) (2020, background): “Cryptography can split the trust in a central party across users (Jayaraman et al., 2018; Bonawitz et al., 2017).”
DoubleSqueeze: Parallel Stochastic Gradient Descent with Double-Pass Error-Compensated Compression (2019, background): “er dierent assumptions. There has been a huge amount number of works studying the implementation of distributed learning from dierent angles. such as dierentially private distributed optimization (Jayaraman et al., 2018; Zhang et al., 2018), adaptive distributed ADMM (Xu et al., 2017), adaptive dist”
Distributed Differentially Private Averaging with Improved Utility and Robustness to Malicious Parties (2020, methodology): “It can be readily used in distributed and federated ML [48, 50] as an alternative to more costly secure aggregation schemes.”
CodedPrivateML: A Fast and Privacy-Preserving Framework for Distributed Machine Learning (2019, background): “In the context of machine learning, DP is mainly used for training when the model parameters are to be released for public use, to ensure that the individual data points from the dataset cannot be identiﬁed from the released model [35]–[41].”
A Comprehensive Survey of Privacy-preserving Federated Learning (2021, unknown): “WHERE RD1 [121, 138, 139] [26, 71] [29, 169] [13, 127, 199] [3, 63, 193] [16, 168, 191] [52, 78, 193] [20, 52, 216] [197, 199] [127, 199] RD2 [61, 129, 194] [39, 137, 215] [157, 198] [125, 164] [10, 70, 172] [109, 155, 174] [110, 172, 228] [57, 81] [56, 165, 222] [30, 198, 224] RD3 [8, 53, 160] [111”
D2P-Fed: Differentially Private Federated Learning With Efficient Communication (2020, methodology): “To improve model utility, secure multiparty computation (SMC) has been used in tandem with DP to reduce noise (Jayaraman et al., 2018; Truex et al., 2019).”
Think Locally, Act Globally: Federated Learning with Local and Global Representations (2020, background): “Differential privacy has also been integrated with deep learning [1], distributed learning [2, 79, 7, 18, 35], and federated learning [29, 73, 28].”
A Scalable Approach for Privacy-Preserving Collaborative Machine Learning (2020, background): “The main application of DP in machine learning is when the model is to be released publicly after training, so that individual data points cannot be backtracked from the released model [25], [26], [27], [28], [29], [30], [31].”