Concealed Decision Tree Computation via Secret Sharing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing decision-tree computation methods require either the user apparatus to share its data with the server apparatus or the server apparatus to share its decision tree, leading to potential leakage of confidential information.
Innovation Solution
A concealed-decision-tree computation system where the user apparatus secret-shares its data and distributes shares among multiple server apparatuses for secret cooperation computation, allowing the user apparatus to obtain computation results without revealing its data or the server's decision tree.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the user apparatus passes its data to the server apparatus for decision-tree computation, then the computation can be performed using the server's decision tree, but the server apparatus will know the user's confidential data
Solution Approach 1:
The user apparatus segments its data into n shares using secret-sharing technology and distributes these shares to n server apparatuses. The server apparatuses then perform decision-tree computation on these shares without reconstructing the original data. This segmentation allows computation to proceed while maintaining data confidentiality, as no single server apparatus possesses the complete data.
Solution Approach 2:
The patent introduces secret shares as an intermediary form between the original data and the computation process. Instead of transmitting raw data to the server, the user apparatus transmits encrypted shares that serve as intermediaries. These shares enable computation but cannot be used to reconstruct the original data by any single server, thus protecting confidentiality while enabling productivity.
2Ease of operation
If the server apparatus passes its decision tree to the user apparatus for computation, then the user can obtain inference results, but the findings of the server apparatus may be leaked
Solution Approach 1:
The server apparatus segments its decision tree into n shares using secret-sharing technology and distributes these shares to n server apparatuses including itself. The user apparatus receives these shares and performs computation locally without obtaining the complete decision tree. This ensures that even if the user apparatus is compromised, the complete decision tree structure remains confidential.
Solution Approach 2:
The patent uses secret shares as intermediaries to enable the user apparatus to perform decision-tree computation without receiving the actual decision tree structure. The shares serve as computational proxies that allow inference operations while preventing reconstruction of the original decision tree, thus protecting the server's proprietary findings.
3Loss of information
If the user apparatus secret-shares data and distributes shares among multiple server apparatuses for secret cooperation computation, then data confidentiality is maintained, but the system complexity increases
Solution Approach 1:
The system segments both data and decision trees into shares that can be independently processed by multiple server apparatuses. This segmentation transforms a complex confidentiality problem into manageable modular units, where each server handles only a portion of the computation, simplifying the overall system architecture while maintaining security.
Data Source
AI summary
A concealed-decision-tree computation system includes a user apparatus and 0th to (n−1)-th server apparatuses, where n is a predetermined positive integer. The user apparatus secret-shares data D into n shares [D]j (j=0, . . . , n−1) and sends the n shares [D]j (j=0, . . . , n−1) to the 0th to (n−1)-th server apparatuses, respectively. The 0th to (n−1)-th server apparatuses use the n shares [D]j (j=0, n−1) to perform secret cooperation computation to obtain n shares [out]0, . . . , [out]n-1 of a value “out” corresponding to the data D in a predetermined decision tree and send the n shares [out]0, . . . , [out]n-1 to the user apparatus. The user apparatus uses at least k shares out of the n received shares [out]0, . . . , [out]n-1 to restore the value “out” corresponding to the data D in the predetermined decision tree, where k is a predetermined integer equal to or smaller than n.