International Association
for Cryptologic Research

<p>Fully Homomorphic Encryption (FHE) is a prevalent cryptographic primitive that allows for computation on encrypted data. In various cryptographic protocols, this enables outsourcing computation to a third party while retaining the privacy of the inputs to the computation. However, these schemes make an honest-but-curious assumption about the adversary. Previous work has tried to remove this assumption by combining FHE with Verifiable Computation (VC). Recent work has increased the flexibility of this approach by introducing integrity checks for homomorphic computations over rings. However, efficient FHE for circuits of large multiplicative depth also requires non-ring computations called maintenance operations, i.e. modswitching and keyswitching, which cannot be efficiently verified by existing constructions. We propose the first efficiently verifiable FHE scheme that allows for arbitrary depth homomorphic circuits by utilizing the double-CRT representation in which FHE schemes are typically computed, and using lattice-based SNARKs to prove components of this computation separately, including the maintenance operations. Therefore, our construction can theoretically handle bootstrapping operations. We also present the first implementation of a verifiable computation on encrypted data for a computation that contains multiple ciphertext-ciphertext multiplications. Concretely, we verify the homomorphic computation of an approximate neural network containing three layers and &gt;100 ciphertexts in less than 1 second while maintaining reasonable prover costs. </p>

Non-Interactive Verifiable Secret Sharing (NI-VSS) is a technique for distributing a secret among a group of individuals in a verifiable manner, such that shareholders can verify the validity of their received share and only a specific number of them can access the secret. VSS is a fundamental tool in cryptography and distributed computing. In this paper, we present an extremely efficient NI-VSS scheme using Zero-Knowledge (ZK) proofs on secret shared data. While prior VSS schemes have implicitly used ZK proofs on secret shared data, we specifically use their formal definition recently provided by Boneh et al. in CRYPTO 2019. The proposed NI-VSS scheme uses a quantum random oracle and a quantum computationally hiding commitment scheme in a black-box manner, which ensures its ease of use, especially in post-quantum threshold protocols. Implementation results further solidify its practicality and superiority over current constructions. With the new VSS scheme, for parameter sets $(n, t)=(128, 63)$ and $(2048, 1023)$, a dealer can share a secret in less than $0.02$ and $2.0$ seconds, respectively, and shareholders can verify their shares in less than $0.4$ and $5.0$ milliseconds. Compared to the well-established Pedersen VSS scheme, for the same parameter sets, at the cost of $2.5\times$ higher communication, the new scheme is respectively $22.5\times$ and $3.25\times$ faster in the sharing phase, and notably needs $271\times$ and $479\times$ less time in the verification. Leveraging the new NI-VSS scheme, we revisit several classic and PQ-secure threshold protocols and improve their efficiency. Our revisions led to more efficient versions of both the Pedersen DKG protocol and the GJKR threshold signature scheme. We show similar efficiency enhancements and improved resilience to malicious parties in isogeny-based DKG and threshold signature schemes. We think, due to its remarkable efficiency and ease of use, the new NI-VSS scheme can be a valuable tool for a wide range of threshold protocols.