AI Model Training via Cross-Device Learning Data Merging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
User devices often have insufficient learning data for training AI models, leading to inconsistent performance across devices, even when used by the same user, due to varying levels of data collection and quality.
Innovation Solution
A method for sharing learning data among multiple external electronic apparatuses, where a processor receives and identifies relevant data from one device to train AI models on another, converting and transmitting the trained models to ensure compatibility and effectiveness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If learning data is collected only from individual user devices, then device-specific personalization is achieved, but data insufficiency leads to inconsistent AI model performance across devices
Solution Approach 1:
The patent combines learning data from multiple external electronic apparatuses (first external electronic apparatus and second external electronic apparatus) into a unified training dataset. The processor identifies corresponding learning data from different devices and merges them to train the AI model, thereby increasing the total volume of learning data while maintaining device-specific characteristics through selective identification and matching of corresponding data across devices.
2Quantity of substance
If learning data is shared across multiple external electronic apparatuses, then data volume increases for better model training, but data identification and matching complexity increases
Solution Approach 1:
The patent uses a correspondence identification mechanism that creates a mapping relationship between learning data from different external electronic apparatuses. The processor identifies corresponding learning data by comparing data characteristics and establishing correspondence relationships, effectively creating a structured copy-linking system that simplifies the matching process across multiple devices without requiring complex manual identification.
3Productivity
If AI models are trained with insufficient learning data, then training time and computational resources are reduced, but model performance and recognition accuracy deteriorate
Solution Approach 1:
The patent merges learning data from multiple external electronic apparatuses to create a comprehensive training dataset. By combining data from the first and second external electronic apparatuses through correspondence identification, the system increases the volume and diversity of training data, thereby improving model performance consistency and recognition accuracy while maintaining efficient training through automated data merging processes.
Data Source
AI summary
An electronic apparatus and a control method thereof are provided. The control method of the electronic apparatus includes receiving, from a first external electronic apparatus and a second external electronic apparatus, a first artificial intelligence model and a second artificial intelligence model used by the first and second external electronic apparatuses, respectively, and a plurality of learning data stored in the first and second external electronic apparatuses, identifying first learning data, which corresponds to second learning data received from the second external electronic apparatus, among learning data received from the first external electronic apparatus, training the second artificial intelligence model used by the second external electronic apparatus based on the first learning data, and transmitting the trained second artificial intelligence model to the second external electronic apparatus.