Local AI Simulation Workflow for Secure User Data Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge in condition-based maintenance of mechanical systems is the difficulty in effectively utilizing AI algorithms with user-specific data due to the risk of disclosing secret information, improper data analysis, and the need for extensive communication and knowledge transfer, leading to inappropriate algorithm selection and inefficient data utilization.
Innovation Solution
A simulation apparatus and method that allows users to verify AI effectiveness using their own data through supervised training, employing a concept of data chunks and a graphical user interface for setting input and output data, enabling in-place analysis and reducing the risk of data exposure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If user data is shared with AI vendors or developers for algorithm training and verification, then AI algorithm effectiveness can be verified, but secret information disclosure risk increases
Solution Approach 1:
The patent introduces a simulation apparatus as an intermediary that receives user data locally and simulates AI algorithm training and verification. This mediator enables effectiveness verification without direct data sharing with external vendors, thus resolving the contradiction between verification reliability and security risk
Solution Approach 2:
The simulation apparatus creates a copy of the AI algorithm training environment on the user's side. By copying the necessary computational functions to the user's device, the system allows verification of AI effectiveness while keeping original data localized, eliminating the need to share sensitive information
2Adaptability or versatility
If extensive communication and knowledge transfer are conducted between users and AI vendors, then appropriate algorithm selection can be achieved, but time and effort are significantly consumed
Solution Approach 1:
The simulation apparatus enables users to independently verify AI algorithm effectiveness using their own data without requiring extensive external assistance. This self-service capability eliminates the need for prolonged communication and knowledge transfer, allowing users to autonomously assess algorithm suitability
Solution Approach 2:
The system performs preliminary AI algorithm verification directly on user devices before deployment decisions are made. By conducting training and evaluation simulations in advance, the system eliminates the need for iterative communication cycles for algorithm selection, saving significant time
3Productivity
If AI algorithms are applied without proper data analysis and verification, then implementation speed is maintained, but algorithm suitability and effectiveness are compromised
Solution Approach 1:
The simulation apparatus performs partial AI training and verification actions locally on user devices before full deployment. By conducting a simplified version of the training process in advance, the system ensures algorithm suitability without requiring complete verification procedures, thus maintaining implementation speed while improving reliability
Data Source
AI summary
A model setting unit executes setting related to a first chunk of input data and a second chunk of training data on a basis of loaded training-purpose data as well as setting related to a third chunk of test input data and a fourth chunk of expected data on a basis of loaded test-purpose data. A model computing unit executes computations of training with use of a machine learning model on a basis of the first chunk and the second chunk, further executes computations of prediction with use of the machine learning model on a basis of results of the training and the third chunk and arithmetically compares results of the prediction and the fourth chunk with each other. The machine learning model after execution of at least part of the computations of training is stored non temporarily.