Faster Fully Homomorphic Encryption: Bootstrapping in Less Than 0.1 Seconds

By Ilaria Chillotti, Nicolas Gama, Mariya Georgieva, Malika Izabachène. 2016

804 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. Hao Chen
3. Wei Dai
4. M. Sadegh Riazi
5. Adrià Gascón
6. Miran Kim
7. Yongsoo Song
8. Blake Pelton

In Computational Theory and Mathematics

1. Kim Laine
2. Hao Chen
3. Wei Dai
4. Miran Kim
5. M. Sadegh Riazi
6. Yongsoo Song
7. Adrià Gascón
8. Blake Pelton

Citations in the corpus

• HEAX: An Architecture for Computing on Encrypted Data (2019, background): “obstacle today to deploy FHE at large-scale is the enormous computation overhead compared to a plaintext counterpart in which data is not kept confidential. MostFHEschemes,i.e.,BGV[11],BFV[31],andTFHE[18] schemes, perform exact computation on encrypted data. A recently proposed FHE scheme called CKK”
• A Review of Homomorphic Encryption Libraries for Secure Computation (2018, methodology): “Brakerski-Gentry-Vaikuntanathan (BGV) [1] and CGGI [2,3] are examples of Fully Homomorphic schemes.”
• Gazelle: A Low Latency Framework for Secure Neural Network Inference (2018, background): “Recent efforts, both in theory and in practice have given us large gains in the performance of several types of PHE schemes and even FHE schemes [4], [7], [8], [15], [19], [32].”
• A Survey on Homomorphic Encryption Schemes (2017, methodology): “HElib is designed using low-level programming, which deals with C++ Yes [Ducas and Micciancio 2015] FFTW TFHE [Chillotti et al. 2017] CGGI16 [Chillotti et al. 2016] C++ Yes [Chillotti et al. 2016] FFTW SEAL [Laine et al. 2017] FV12 [Fan and Vercauteren 2012b] C++ Yes [Chen et al. 2017] No external…”
• Bootstrapping for Approximate Homomorphic Encryption (2018, background): “[17] obtained a speed up to less than 0.”
• TAPAS: Tricks to Accelerate (encrypted) Prediction As a Service (2018, methodology): “Since then several other schemes have been proposed Gentry et al. (2012, 2013); Brakerski and Vaikuntanathan (2014); Ducas and Micciancio (2015); Chillotti et al. (2016).”
• Fast Homomorphic Evaluation of Deep Discretized Neural Networks (2018, methodology): “Hence, there have been numerous works on trying to obtain more efficient bootstrappings [AP13, AP14, DM15, CGGI16b, CGGI17], and on trying to minimize the number of bootstrappings required for evaluating a circuit [LP13, PV16, BLMZ17].”
• A Pragmatic Introduction to Secure Multi-Party Computation (2019, background): “Although there has been much recent interest in implementing FHE schemes Gentry and Halevi (2011), Halevi and Shoup (2015), and Chillotti et al. (2016), building secure, deployable, scalable systems using FHE remains an elusive goal.”
• Efficient Multi-Key Homomorphic Encryption with Packed Ciphertexts with Application to Oblivious Neural Network Inference (2019, methodology): “However, we stress that our assumption is no stronger than the same assumption in HE schemes [16, 17, 22, 27] requiring either bootstrapping or relinearization of ciphertexts.”
• HEAX: High-Performance Architecture for Computation on Homomorphically Encrypted Data in the Cloud (2019, methodology): “, the BGV [11], the BFV [31], and the TFHE [18] perform exact computation on encrypted data.”
• PrivEdge: From Local to Distributed Private Training and Prediction (2020, background): “FHE is computationally expensive and, despite recent advances [34]–[37], training a machine learning classifier (e.”
• Improving TFHE: faster packed homomorphic operations and efficient circuit bootstrapping (2017, methodology): “Like in [12], a TLWE ciphertext are not viewed as a fixed values but as a distribution over the Ω-probability space, which is the product of the probability spaces of each fresh TLWE sample and of possibly randomized vector coefficients ri in Z or R.”
• Labeled PSI from Fully Homomorphic Encryption with Malicious Security (2018, background): “Fully Homomorphic Encryption (FHE) is a form of encryption that allows arbitrary (Boolean or arithmetic) circuits to be evaluated directly on encrypted data without requiring access to the decryption key [49,24,9,10,7,22,8,26,17,13].”
• TFHE: Fast Fully Homomorphic Encryption Over the Torus (2019, methodology): “This paper is the complete and extended version of the two papers [22,24] published by the same authors at Asiacrypt 2016 and Asiacrypt 2017, respectively.”
• Security and Privacy Issues in Deep Learning (2018, methodology): “[124] supports operations on binary data, both utilized the concept of Binary Neural Networks (BNNs) [125].”