Skip to main content

​
Pinocchio: Nearly Practical Verifiable Computation

The Pinocchio system presents a significant advancement in the field of verifiable computation, demonstrating a method for creating computationally sound and ne

​
Abstract

The Pinocchio system presents a significant advancement in the field of verifiable computation, demonstrating a method for creating computationally sound and nearly practical proofs for arbitrary computations. It allows a prover to generate a proof that a computation was performed correctly, and a verifier can check this proof efficiently without re-executing the entire computation. Pinocchio builds upon techniques from homomorphic encryption and zero-knowledge proofs, specifically utilizing the Groth-Sahai proof system and incorporating techniques to reduce the proof size and verification time. The system achieves efficiency by offloading the bulk of the computation verification to the prover, requiring the verifier only to perform a small, constant number of operations regardless of the computation’s complexity. While not fully “zero-knowledge” in the strictest sense, it offers a compelling trade-off between proof generation cost and verification efficiency, making verifiable computation more accessible for practical applications, including blockchains and secure multi-party computation. Authors: Bryan Parno, Jon Howell, Craig Gentry, Mariana Raykova Publication: IACR ePrint Archive Publication date: 2013-01-01

​
Key findings

  • Pinocchio enables efficient generation and verification of proofs for arbitrary computations.
  • It allows a verifier to confirm computation correctness without re-executing the entire process.
  • The system leverages techniques from homomorphic encryption and zero-knowledge proofs for efficiency.
  • Pinocchio achieves nearly practical performance by minimizing verifier workload, making verifiable computation more accessible.

​
Citation

Bryan Parno, Jon Howell, Craig Gentry, Mariana Raykova (2013). Pinocchio: Nearly Practical Verifiable Computation. IACR ePrint Archive. https://eprint.iacr.org/2013/279
Canonical knowledge ID: research:pinocchio-nearly-practical-verifiable-computation