Efficient Multiparty Protocols Using Circuit Randomization

By Donald Beaver. 1991

How researchers see this paper

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

Citations in the corpus

Secure Evaluation of Quantized Neural Networks (2019, neutral): “one of the parties has a different role in the computation, which simplifies many of the expensive procedures like generating random multiplication triples [6].”
Privacy-Preserving Distributed Linear Regression on High-Dimensional Data (2017, methodology): “It builds upon techniques based on precomputation of multiplicative triples [2], as the secret sharing techniques mentioned above, and relies on an initializer that aids in the computation.”
Prio: Private, Robust, and Scalable Computation of Aggregate Statistics (2017, methodology): “Then, the servers can execute Beaver’s multiparty computation protocol (Section C.2) to evaluate the circuit using the M client-provided multiplication triples.”
Chameleon: A Hybrid Secure Computation Framework for Machine Learning Applications (2018, positive): “1 , and they wish to obtain shares ⟨z⟩B 0 and ⟨z⟩B 1 respectively. As shown in [37], the most efficient method for evaluating AND gates in the GMW protocol is based on Beaver’s multiplication triples [11]: Multiplication triples are random shared-secrets a, b, and c such that ⟨c⟩B 0 ⊕⟨c⟩B 1 = (⟨a⟩B ”
More efficient oblivious transfer and extensions for faster secure computation (2013, methodology): “An AND gate in the GMW protocol can be computed efficiently using the multiplication triple functionality [3]: the parties hold no input, and the functionality chooses random bits a0, a1, b0, b1, c0, c1 under the constraint that c0 ⊕ c1 = (a0⊕a1)(b0⊕ b1).”
Practical Covertly Secure MPC for Dishonest Majority - Or: Breaking the SPDZ Limits (2013, background): “In [12, 14] the need to amortize the cost of the expensive zero-knowledge proofs meant that the players on each iteration of the offline protocol executed a large computation, which produced a large number of multiplication triples [4] (in the millions).”
Zero-Knowledge Proofs on Secret-Shared Data via Fully Linear PCPs (2019, background): “, in the form of multiplication triples [12]) to enable MPC with dishonest majority.”
High-Throughput Secure Three-Party Computation for Malicious Adversaries and an Honest Majority (2017, methodology): “Our approach to achieving malicious security follows the Beaver multiplication triple approach [2] used in [19,9,5] (and many follow-up works).”
Oblivious Neural Network Predictions via MiniONN Transformations (2017, methodology): “Recall that we introduce a precomputation phase to generate dotproduct triplets, which are similar to themultiplication triplets used in secure computations [8].”
Secure and Efficient Federated Transfer Learning (2019, background): “In contrast, each multiplication of shared values requires a unique multiplication triple [12].”
Perfectly-Secure MPC with Linear Communication Complexity (2008, methodology): “Furthermore, all but the perfectly secure protocol use circuit randomization [ Bea91a ], which reduces the multiplication of two shared values to two reconstructions, given a precomputed sharing of a random multiplication triple (a, b, c )w ithc = ab. Such triples can be non-robustly generated and c”
SoK: General Purpose Compilers for Secure Multi-Party Computation (2019, methodology): “Arithmetic-circuit GMW-based protocols usually generate some form of “Beaver Multiplication Triples” (secret shares of random triples (a, b, a · b), where a, b are field elements) [10] that are used as masks in the online phase.”
ABY2.0: Improved Mixed-Protocol Secure Two-Party Computation (2020, methodology): “Before going into the details, we present a high-level overview of our scheme and a side-by-side comparison with the well-known Beaver’s circuit randomization technique [8].”
Trident: Efficient 4PC Framework for Privacy Preserving Machine Learning (2019, methodology): “Offline-online paradigm: To improve efficiency, a class of MPC protocols operates in the offline-online paradigm [28].”