(Leveled) Fully Homomorphic Encryption without Bootstrapping

By Zvika Brakerski, Craig Gentry, Vinod Vaikuntanathan. 2014

3,339 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Theory and Mathematics, Information Systems, 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 Artificial Intelligence

1. Kim Laine
2. Dan Boneh
3. Shai Halevi
4. Wei Dai
5. Craig Gentry
6. Hao Chen
7. Vinod Vaikuntanathan
8. Brent Waters

In Computational Theory and Mathematics

1. Shai Halevi
2. Kim Laine
3. Hao Chen
4. Craig Gentry
5. Vinod Vaikuntanathan
6. Dan Boneh
7. Chris Peikert
8. Miran Kim

In Information Systems

1. Wei Dai
2. Kim Laine
3. Berk Sunar
4. Dan Boneh
5. M. Sadegh Riazi
6. Yarkin Doröz
7. Blake Pelton
8. Brent Waters

In Privacy & Security

1. Shai Halevi
2. Srinath T. V. Setty
3. Kim Laine
4. Hao Chen
5. Vinod Vaikuntanathan
6. Kristin E. Lauter
7. Adrià Gascón
8. Peter Kairouz

Citations in the corpus

• Advances and Open Problems in Federated Learning (2019, background): “Different flavours of HE exist, ranging from general fully homomorphic encryption (FHE) [176] to the more efficient leveled variants [79, 160, 80, 112], for which several implementations exist [3, 350, 4].”
• DISCO: Dynamic and Invariant Sensitive Channel Obfuscation for deep neural networks (2020, background): “For the former, distributed learning techniques such as federated learning [6, 7] and split learning [8, 9] are used, where clients communicate with a centralized server using weights and activations and the latter relies on homomorphic encryption [10, 11] and secure enclaves [12, 13].”
• HEAX: An Architecture for Computing on Encrypted Data (2019, neutral): “We evaluate our design on a wide range of FHE parameters: from ciphertext polynomial size ( n ) of 2 12 and 109-bit ciphertext modulus ( ⌊ log qp ⌋ + 1) to 2 14 with 438-bit ciphertext modulus.”
• Private Stateful Information Retrieval (2018, neutral): “Using leveled fully homomorphic encryption [13] and Reed-Muller codes [42, 46], Hamlin et al.”
• PIR with Compressed Queries and Amortized Query Processing (2018, background): “BGV [16], FV [36]) for values of j ∈ [0, n − 1] to obtain the n ciphertexts.”
• XPIR : Private Information Retrieval for Everyone (2016, methodology): “The Homomorphic-Encryption Library of Halevi and Shoup [45] implements the encryption scheme of Brak-erski, Gentry and Vaikuntanathan [46].”
• CryptoDL: Deep Neural Networks over Encrypted Data (2017, background): “Recent advances in fully homomorphic encryption (FHE) enable a limited set of operations to be performed on encrypted data [11,55].”
• Gazelle: A Low Latency Framework for Secure Neural Network Inference (2018, methodology): “Examples of depth-bounded homomorphic encryption scheme (called leveled homomorphic encryption or LHE) are the family of lattice-based schemes such as the Brakerski-GentryVaikuntanathan [4] scheme and its derivatives [6], [13].”
• A Survey on Homomorphic Encryption Schemes (2017, neutral): “Brakerski et al. [Brakerski et al. 2014] also showed that their scheme with boot-strapping still provides better performance than the one without bootstrapping and also suggested the batching as an optimization.”
• Private genome analysis through homomorphic encryption (2015, methodology): “In particular, two techniques are used for noise management of SHE: one is the modulus-switching technique introduced by Brakerski, Gentry and Vaikuntanathan [21], which scales down a ciphertext during every multiplication operation and reduces the noise by its scaling factor.”
• Secure Linear Regression on Vertically Partitioned Datasets (2016, neutral): “While there are cryptographic techniques such as secure multiparty computation (MPC) [5,17,40] and fully homomorphic encryption (FHE) [7,8,15] that provide general solutions for the problem of computation on private data, using these techniques in a black-box way rarely scales to problem sizes commo”
• HOP: Hardware makes Obfuscation Practical (2017, methodology): “Over a 1 GByte dataset, we evaluate this function with a 10level FHE circuit, which gives an FHE polynomial dimension (n) of ∼ 8192 and ciphertext space q of ∼ 2128 (using terminology from [9]).”
• Efficient Pseudorandom Correlation Generators: Silent OT Extension and More (2019, methodology): “For these constructions we use HSS based on ring-LWE [23, 31, 22] and the BGN (pairingbased) cryptosystem [13, 18, 16].”
• Homomorphic Encryption for Arithmetic of Approximate Numbers (2017, methodology): “Technically this procedure is similar to the modulus-switching algorithm [5], but it has a completely different role in our construction.”
• Adding Distributed Decryption and Key Generation to a Ring-LWE Based CCA Encryption Scheme (2019, methodology): “To understand prior techniques for distributed decryption in the Ring-LWE setting, we will consider a simple BGV style Ring-LWE [6] ciphertext, under a secret key which is distributed with respect to a full threshold access structure.”