Secure Quantile Bucketing Using Secret-Shared Count Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for quantile bucketing on joint data between parties require high communicative complexity and are difficult to scale, especially when data privacy and security are paramount, as they often involve sorting and sharing raw data values.
Innovation Solution
Precomputing a secret shared count lookup table to determine bucket thresholds, allowing secure comparisons and assignments without revealing actual data values, thus reducing communicative complexity and enabling scalable quantile bucketing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If prior sorting-based techniques are used for secure quantile bucketing, then data privacy is maintained, but communicative complexity grows on the order of n log n
Solution Approach 1:
The patent precomputes a count lookup table that stores cumulative counts of data values before the actual bucketing operation. This preliminary computation allows the bucketing process to directly query pre-calculated thresholds rather than performing complex sorting and communication protocols during the main operation, reducing communicative complexity from O(n log n) to O(n) while preserving data privacy through secure multi-party computation
Solution Approach 2:
The patent creates a count lookup table that is a simplified representation or copy of the full data distribution information. Instead of requiring parties to communicate and compare actual data values during bucketing, they use this precomputed table which captures the essential distribution characteristics, enabling efficient threshold determination without revealing underlying private data
2Measurement precision
If sorting is performed on joint data, then accurate quantile bucketing is achieved, but duplication issues arise and scalability is limited
Solution Approach 1:
The count lookup table is precomputed to store cumulative frequency information that directly enables accurate quantile calculation. This allows the system to determine bucket thresholds by simple table lookups and comparisons rather than performing resource-intensive sorting operations on the actual joint data, achieving both accuracy and scalability
Solution Approach 2:
The patent transforms the problem from operating on raw data values to operating on aggregated count parameters. By changing the representation from individual data points to cumulative counts in the lookup table, the system achieves linear-time bucketing that scales efficiently while maintaining the precision needed for accurate quantile determination
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing secure quantile bucketing. One of the methods includes for a first party and a second party, performing secure quantile bucketing on a joint dataset comprising first data of the first party and second data of the second party, the performing comprising: precomputing a secret shared count lookup table for the joint dataset; determine secret shares of bucket thresholds for each bucket interval; assign data points to the buckets based on the secret shared bucket thresholds; and provide the first party and the second party with an output comprising a list of bucket thresholds in secret shared form and an label per data point in secret shared form identifying the bucket assigned to each data point.