Storage Device Insight Sharing Without Raw Data Exposure
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data storage devices face challenges in optimizing performance and reliability while preserving data privacy and security, particularly in machine learning and AI model training, due to restrictions on data sharing and privacy regulations.
Innovation Solution
Implementing a privacy-preserving information-sharing method between data storage devices through federated learning protocols, allowing the sharing of data insights without exposing individual device data, by using predictive models trained on collective data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data from multiple storage devices is shared to improve ML algorithm performance, then the performance and reliability of storage devices is improved, but data privacy and security are compromised
Solution Approach 1:
The patent extracts only the necessary information (predictive conclusions and parameters gradients) from the raw data while leaving the actual data behind. Each storage device shares derived insights rather than raw data, allowing performance improvement without compromising data privacy. This extraction of useful information while separating it from the original data source directly resolves the contradiction between sharing data for improvement and protecting data privacy.
Solution Approach 2:
The patent introduces an intermediary mechanism where storage devices share predictive conclusions and model parameters gradients rather than raw data. This intermediary layer of derived information enables collaboration between storage devices while maintaining data privacy, as the intermediary prevents direct exposure of raw data between devices.
2Object-affected harmful factors
If data is separated to ensure privacy and security, then data privacy is improved, but the ability to optimize performance through collective learning is reduced
Solution Approach 1:
The patent creates copies of the predictive model and parameters gradients that can be shared between storage devices without copying the actual data. Each device maintains its own data while sharing copies of the trained model parameters, enabling performance optimization through collective learning while preserving data separation and privacy.
Solution Approach 2:
The patent changes the parameters of the predictive model through collaborative learning while keeping the data parameters (raw data) separate and private. By updating model parameters based on aggregated insights from multiple devices without moving the actual data, the system achieves performance optimization while maintaining data privacy and security.
3Loss of information
If raw data is shared between storage devices, then collective data analysis is improved, but information security and data isolation are compromised
Solution Approach 1:
The patent extracts only the essential predictive conclusions and model parameters from the raw data for sharing purposes. By taking out only the derived insights and leaving the raw data isolated in each storage device, the system enables collective data analysis capability while maintaining information security and data isolation.
Data Source
AI summary
Data privacy and fulfilling security limitations are ensured during ML algorithm and AI model training by forcing the distinct separation of stored data of each data storage device and preventing the allowance of information sharing between other data storage devices. Specifically, a privacy-preserving information-sharing method is implemented between data storage devices in a joint system. The data of each storage device is not exposed to other storage devices in the joint system. Instead, predictive conclusions based on statistical and ML analysis derived from the collective data of all the storage devices is observed by each storage device. Thus, by allowing the sharing of data insights between storage devices without exposing the data of each storage device to other storage devices, performance and reliability of a storage device is improved.