Can homomorphic encryption be practical?

By Kristin E. Lauter, Michael Naehrig, Vinod Vaikuntanathan. 2011

1,184 citations recorded by the source corpus.

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

1. Kim Laine
2. Shai Halevi
3. Miran Kim
4. Craig Gentry
5. Dan Boneh
6. Hao Chen
7. Frederik Vercauteren
8. Wei Dai

In Computational Theory and Mathematics

1. Shai Halevi
2. Kim Laine
3. Miran Kim
4. Hao Chen
5. Craig Gentry
6. Michael Naehrig
7. Kristin E. Lauter
8. Frederik Vercauteren

Citations in the corpus

• Privacy-Preserving Ridge Regression on Hundreds of Millions of Records (2013, methodology): “[50] mention using FHE for regression, but do not quantify performance.”
• Crypto-Nets: Neural Networks over Encrypted Data (2014, background): “Efficiency can be increased by restricting to somewhat homomorphic schemes and by operating on integers instead of bits, see Lauter et al. (2011).”
• A Survey on Homomorphic Encryption Schemes (2017, methodology): “Actually, the idea of homomorphic evaluation of AES is first discussed in [Naehrig et al. 2011] with the following scenario.”
• Fully homomorphic SIMD operations (2012, methodology): “Evaluation of AES circuits using FHE operations has been mentioned as a possible usage scenario in [13].”
• Fully Homomorphic Encryption with Polylog Overhead (2012, methodology): “Ultimately, the Add,Mult and Permute operations can all be accomplished with Õ(λ) computation by building on the recent Brakerski-Gentry-Vaikuntanathan (BGV) “FHE without bootstrapping” scheme [5], which builds on prior work by Brakerski and Vaikuntanathan and others [7–9].”
• Efficient Secure Matrix Multiplication Over LWE-Based Homomorphic Encryption (2016, methodology): “According to the experimental estimation of [5], we may take U = 2tσ(2) √ n in practice.”
• (Leveled) fully homomorphic encryption without bootstrapping (2012, neutral): “construction based on ideal lattices [8], there are by now a number of such schemes in the literature [6, 20, 10, 5, 4, 13], all of which are based on lattices (either directly or implicitly).”
• SHIELD: Scalable Homomorphic Implementation of Encrypted Data-Classifiers (2015, methodology): “Applications analyzed in this paper were primarily inspired from [9], [21].”
• Homomorphic Evaluation of the AES Circuit (2012, background): “Some of these works were accompanied by implementation [19, 10, 6, 20, 14, 7], but all the implementations so far were either “proofs of concept” that can compute only one basic operation at a time (at great cost), or special-purpose implementations limited to evaluating very simple functions.”
• Privacy-preserving matrix factorization (2013, methodology): “Non-linear computation through fully homomorphic encryption [16] may be used to reduce the workload on the CSP compared to garbled circuits, but current FHE schemes [39, 20] for simpler algebraic computations are not as efficient as garbled circuit approaches [50].”
• An analysis of security issues for cloud computing (2013, background): “The authors in [77] provided some real-world cloud applications where some basic homomorphic operations are needed.”
• Accelerating Fully Homomorphic Encryption over the Integers with Super-size Hardware Multiplier and Modular Reduction (2013, neutral): “024 seconds [12].”
• A Survey of Provably Secure Searchable Encryption (2014, neutral): “Since the first FHE scheme, researchers have tried to make the schemes more efficient, but still no practical construction has been proposed [Naehrig et al. 2011].”
• Secure Outsourced Matrix Computation and Application to Neural Networks (2018, neutral): “This technology has great potentials in many real-world applications such as statistical testing, machine learning, and neural networks [24, 30, 31, 40].”
• Public Key Compression and Modulus Switching for Fully Homomorphic Encryption over the Integers (2012, background): “A partial implementation is described in [12], without the fully homomorphic capability.”
• Modular Hardware Architecture for Somewhat Homomorphic Function Evaluation (2015, background): “…that homomorphic evaluation of a block cipher decryption is required to reduce the network communication to the data size following the proposal in [32]: data is encrypted with a block cipher by the user and then the server decrypts the ciphertext by evaluating the block cipher homomorphically by…”