2021/07/09 by Sanket Kanjalkar, Kanjalkar, Sanket, Ye Zhang +5 · 1 citation
Computer Science · #Cryptography and Data Security #Cryptography and Security (cs.CR) #FOS: Computer and information sciences #Privacy-Preserving Technologies in Data #Security and Verification in Computing
paper · pdf · doi:10.48550/arxiv.2107.04248
openalex publication_date 2021/07/09 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
In recent years, multiparty computation as a service (MPCaaS) has gained\npopularity as a way to build distributed privacy-preserving systems. We argue\nthat for many such applications, we should also require that the MPC protocol\nis publicly auditable, meaning that anyone can check the given computation is\ncarried out correctly -- even if the server nodes carrying out the computation\nare all corrupt. In a nutshell, the way to make an MPC protocol auditable is to\ncombine an underlying MPC protocol with verifiable computing proof (in\nparticular, a SNARK). Building a general-purpose MPCaaS from existing\nconstructions would require us to perform a costly "trusted setup" every time\nwe wish to run a new or modified application. To address this, we provide the\nfirst efficient construction for auditable MPC that has a one-time universal\nsetup. Despite improving the trusted setup, we match the state-of-the-art in\nasymptotic performance: the server nodes incur a linear computation overhead\nand constant round communication overhead compared to the underlying MPC, and\nthe audit size and verification are logarithmic in the application circuit\nsize. We also provide an implementation and benchmarks that support our\nasymptotic analysis in example applications. Furthermore, compared with\nexisting auditable MPC protocols, besides offering a universal setup our\nconstruction also has a 3x smaller proof, 3x faster verification time and\ncomparable prover time.\n