Learning model generation method, program, storage medium, and learned model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The discovery of a preferable combination of water-repellent agents requires repetitive tests and evaluations, resulting in a significant burden in terms of time and cost.
Innovation Solution
A learning model generation method that uses a computer to generate a model for evaluating an article with a surface-treating agent fixed onto a base material by obtaining teacher data, learning from it, and generating a model that can output evaluations for unknown input information, thereby reducing the time and cost of evaluations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional test and evaluation methods are used to discover preferable combinations of water-repellent agents, then reliable evaluation results are obtained, but significant time and cost are required
Solution Approach 1:
The invention creates a virtual copy of the physical evaluation process through a learning model. The model is trained on teacher data obtained from actual water-repellency tests, then uses this learned knowledge to predict evaluations for new combinations without conducting physical tests. This copying approach maintains evaluation reliability while eliminating time-consuming repeated testing.
Solution Approach 2:
The invention performs preliminary action by collecting and storing teacher data from initial tests, then training the learning model in advance. Once trained, the model can rapidly evaluate new combinations without requiring new physical tests. This preliminary preparation of the evaluation system enables quick subsequent assessments.
2Reliability
If conventional test and evaluation methods are used to discover preferable combinations of water-repellent agents, then reliable evaluation results are obtained, but significant cost is required
Solution Approach 1:
The learning model creates a virtual evaluation system that copies the functionality of physical testing. Once the model is trained using teacher data from initial tests, it can perform unlimited evaluations at minimal computational cost, eliminating the need for repeated expensive physical tests while maintaining reliable evaluation results.
3Productivity
If a learning model is generated to enable computer-based evaluation, then evaluation time and cost are reduced, but the model requires training data and computational resources
Solution Approach 1:
The learning model serves as an intermediary between the complex physical testing process and the simple prediction task. It absorbs the complexity of understanding water-repellency mechanisms during training, then provides simple, fast predictions. This intermediary approach enables high productivity while managing complexity within the model structure.
Data Source
AI summary
A learning model generation method may include obtaining, by a processor, as teacher data, information including at least first base material information regarding a first base material, first treatment agent information regarding a first surface-treating agent, and a first evaluation of a first article; learning, by the processor, based on the teacher data; and generating, by the processor, a learning model based on the learning. A second article may be obtained by fixing a second surface-treating agent onto a second base material. The learning model may be configured to receive input information, which is different from the teacher data, as an input, and output a second evaluation of the second article. The input information may include at least second base material information regarding the second base material, and second treatment agent information regarding the second surface-treating agent.