Split Algorithm Computation with Beaver Sets for Privacy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies face challenges in securely and efficiently performing complex operations on encrypted data and algorithms, particularly in multiparty computation scenarios where privacy of both data and algorithms must be maintained, often requiring numerous communications and being limited by the complexity of operations like multiplication.
Innovation Solution
The system employs a method to split algorithms and data into subsets, using Beaver sets to reduce communication hops and enable faster computations by preprocessing, and applies cryptographic techniques like secure multi-party computation and homomorphic encryption to maintain privacy, allowing operations on encrypted data without revealing proprietary information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If secure multi-party computation is used to maintain privacy of data and algorithms, then privacy protection is improved, but communication overhead and computation time increase
Solution Approach 1:
The algorithm is divided into multiple shares distributed across different parties. Each party holds only a portion of the algorithm and data, allowing local computation without revealing the complete algorithm or data to others. This segmentation enables privacy preservation while reducing the communication burden compared to transmitting full encrypted datasets.
Solution Approach 2:
Beaver triples are pre-computed and distributed before the actual computation. These pre-computed values enable parties to perform multiplication operations locally without requiring extensive communication during the computation process, thereby reducing real-time communication overhead while maintaining security.
2Reliability
If algorithms are split into shares for secure computation, then privacy of algorithm is improved, but computation complexity increases
Solution Approach 1:
The algorithm is segmented into shares that are distributed across multiple parties. Each party performs computation on their local share without needing to understand the complete algorithm. This segmentation simplifies individual computation tasks while maintaining overall algorithm privacy through the distributed structure.
Solution Approach 2:
Beaver triples act as intermediaries that facilitate secure multiplication operations between algorithm shares. These pre-computed intermediary values enable parties to perform complex operations locally using simpler operations on the shares, reducing the computational complexity burden on each individual party while maintaining algorithm privacy.
3Reliability
If homomorphic encryption is applied to perform operations on encrypted data, then data privacy is improved, but computational efficiency decreases
Solution Approach 1:
Data is segmented into shares distributed across multiple parties rather than encrypting the entire dataset with homomorphic encryption. Each party performs computation on their local share, and results are combined at the end. This approach maintains data privacy while significantly improving computational efficiency compared to homomorphic encryption of large datasets.
Solution Approach 2:
Beaver triples serve as intermediaries that enable efficient computation on shared data. By pre-computing and distributing these intermediary values, parties can perform multiplication and other complex operations locally on data shares without requiring homomorphic encryption, thereby maintaining privacy while improving computational efficiency.
Data Source
AI summary
The disclosed concepts achieve privacy for data operated on by an algorithm in an efficient manner A method includes receiving a first algorithm subset, receiving a second algorithm subset, generating two shares of a first mathematical set based on the first algorithm subset and transmitting the two shares of the first mathematical set from a first entity to a second entity. The method can include generating two shares of a second mathematical set based on the second algorithm subset, transmitting the two shares of the second mathematical set from the second entity to the first entity, receiving first split data subset of a full data set and receiving a second split data subset of the full data set. The system, based on these subsets of data, generates a first output subset and a second output subset which are combined for the final output.