Logsum using Garbled Circuits.

Area of Science:

• Cryptography
• Secure Multiparty Computation
• Privacy-Preserving Technologies

Background:

• Secure multiparty computation (SMC) enables joint function evaluation on private data.
• Fully homomorphic encryption (FHE) is theoretically capable but computationally impractical for SMC.
• Garbled Circuits (GCs) offer a practical alternative for secure function evaluation, with recent advancements.

Purpose of the Study:

• To develop an efficient and accurate method for computing the logsum operation within the Garbled Circuits framework.
• To leverage recent breakthroughs in Garbled Circuits design and implementation for privacy-preserving computations.

Main Methods:

• The proposed technique approximates the logsum operation using a piecewise linear function.
• This approximation is implemented using Garbled Circuits, a cryptographic technique for secure function evaluation.
• The method focuses on assembling necessary computational blocks for accuracy and speed.

Main Results:

• The technique achieves a highly accurate approximation of the logsum operation.
• The implementation demonstrates very fast execution times, making it practical for real-world applications.
• The method effectively balances accuracy and computational efficiency in secure computation.

Conclusions:

• The developed Garbled Circuit-based logsum computation offers a practical solution for privacy-preserving data analysis.
• This approach overcomes the computational limitations of fully homomorphic encryption for certain functions.
• The piecewise linear approximation within Garbled Circuits provides an efficient and accurate method for secure logsum evaluation.