Additional Learning Model Updates for Unknown Disturbance Events
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies require manual intervention for updating models in manufacturing devices due to frequent material or environmental changes, making it difficult to manage unexpected disturbances efficiently.
Innovation Solution
An additional learning device that autonomously acquires and processes data from a control object, determines unknown explanatory variables, and updates the model by correlating these variables with corresponding objective variables, thereby eliminating the need for manual intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual decision-making is used to update models when material or environmental changes occur, then model accuracy can be maintained, but productivity decreases due to frequent manual interventions
Solution Approach 1:
The system performs self-learning by automatically detecting unknown explanatory variables through the determination unit and updating models through additional learning without requiring manual intervention. The acquisition unit continuously gathers data, the determination unit identifies unknown variables, and the additional learning unit updates the model autonomously, enabling the system to adapt to material or environmental changes while maintaining productivity.
2Reliability
If manual update decisions are required for each model change, then model updates can be controlled, but loss of time increases due to repeated human intervention
Solution Approach 1:
The system implements a feedback mechanism where the determination unit continuously monitors for unknown explanatory variables, and when detected, triggers automatic additional learning to update the model. This closed-loop feedback system eliminates manual decision-making time by automatically responding to changes in material or environmental conditions while maintaining controlled model updates.
3Adaptability or versatility
If the model is updated frequently to adapt to changes, then adaptability improves, but device complexity increases due to additional learning mechanisms
Solution Approach 1:
The system segments the model updating process into distinct functional units: an acquisition unit for data collection, a determination unit for identifying unknown variables, and an additional learning unit for model updates. This segmentation allows the system to achieve high adaptability through targeted additional learning only when unknown variables are detected, rather than continuously updating the entire model, thus managing complexity effectively.
Data Source
AI summary
The disclosure provides an additional learning device, an additional learning method, and a storage medium. The additional learning device includes: an acquisition unit acquires information based on data observed in a control object which is controlled by an instruction in accordance with an output from a previously learned model outputting an objective variable with respect to an explanatory variable. A determination unit determines whether or not the explanatory variable corresponding to the information acquired by the acquisition unit exists as learning data of the model, that is, whether or not the explanatory variable is unknown. When the explanatory variable is determined to be unknown by the determination unit, an additional learning unit acquires the objective variable corresponding to the unknown explanatory variable, correlates the unknown explanatory variable and the acquired objective variable, and additionally learns the model. The model is updated with regard to unknown disturbance events without requiring manpower.