Secure Multiparty Computation for Privacy-Preserving Data Mining

By Yehuda Lindell, Benny Pinkas. 2009

791 citations recorded by the source corpus.

In this paper, we survey the basic paradigms and notions of secure mul- tiparty computation and discuss their relevance to the fleld of privacy-preserving data mining. In addition to reviewing deflnitions and constructions for secure mul- tiparty computation, we discuss the issue of e-ciency and demonstrate the di-cul- ties involved in constructing highly e-cient protocols. We also present common errors that are prevalent in the literature when secure multiparty computation techniques are applied to privacy-preserving data mining. Finally, we discuss the relationship between secure multiparty computation and privacy-preserving data mining, and show which problems it solves and which problems it does not.

Subfields: Artificial Intelligence, Computational Theory and Mathematics

How researchers see this paper

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

In Artificial Intelligence

In Computational Theory and Mathematics

Citations in the corpus

A Comprehensive Comparison of Multiparty Secure Additions with Differential Privacy (2015, methodology): “We use the formal security notion to present the secure multiparty computation protocols [30], [39], [55].”
Hacking smart machines with smarter ones: How to extract meaningful data from machine learning classifiers (2013, methodology): “…through heuristic values (e.g., selecting the values that minimize the utility loss of the data), cryptographic protocols (e.g., via Secure Multiparty Computation [44]), or reconstruction-based techniques (e.g., strategy aimed at reconstructing the original data distribution using randomized data).”
Differentially Private Distributed Online Learning (2018, neutral): “, [1], [2], [3], [4]).”
Distributed Data Mining with Differential Privacy (2011, methodology): “The adaptive semi-honest adversary model proposed in [12] will be used in our system.”
Secure multiparty aggregation with differential privacy: a comparative study (2013, methodology): “This problem is typically formulated as a secure multiparty computation problem (SMC), in which parties wish to jointly compute a function of their private data inputs [2,8,9,30,36,37].”
Privacy-Preserving Record Linkage (2010, neutral): “, [27, 26] for an overview.”
Computational Differential Privacy from Lattice-Based Cryptography (2017, background): “In contrast to common secure multi-party techniques (see [18], [23]), this protocol requires each user to send only one message per query to the analyst.”
Privacy-Preserving Visual Learning Using Doubly Permuted Homomorphic Encryption (2017, background): “The homomorphic cryptosystems also enable a variety of privacy-preserving machine learning and data mining [7, 14, 25, 39, 43, 44, 54, 57, 61] because they can make some weights perfectly secret by encrypting them.”
Hiding in the Crowd: A Massively Distributed Algorithm for Private Averaging with Malicious Adversaries (2018, background): “…Graepel et al., 2012; Aslett et al., 2015) require the existence of a trusted third party, secure multi-party computation techniques (Yao, 1982; Lindell and Pinkas, 2009) are generally intractable when the number of parties is large, and exchanging noisy sketches of the data through (local)…”
Distributed Learning without Distress: Privacy-Preserving Empirical Risk Minimization (2018, background): “Lindell and Pinkas [31] proposed one of the earliest approaches to use MPC for private data mining, which was followed by several works considering different adversarial models or applications [55, 50, 33, 42].”
A taxonomy of privacy-preserving record linkage techniques (2013, neutral): “(ii) Secure multi-party computation ( SMC ): The basic idea of SMC is that a computation is secure if at the end of the computation no party knows anything except its own input and the ﬁnal results of the computed function [82,83,93].”
Federated Neural Architecture Search (2020, background): “Nevertheless, the distribution vectors can be further encrypted through secure multiparty computation [23].”
Oblivious Multi-Party Machine Learning on Trusted Processors (2016, background): “Lindell and Pinkas [36] survey these techniques with respect to data mining tasks including machine learning.”
SoK: General Purpose Compilers for Secure Multi-Party Computation (2019, background): “broad potential for practical applications, ranging from general computations of secure statistical analysis [22], [23], [25], [52], [53], [54], [55], [96], to more domainspecific uses like financial oversight [2], [21], [26], [58], biomedical computations [35], [32], [75], [84], [82], [85], [132] a”
Security and Privacy Issues in Deep Learning (2018, background): “Secure multi-party computation (SMC) occurs when a set of parties like silos, with private inputs wish to compute some joint function of their inputs [122].”