Teaching Device for Machine Learning Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Machine learning models often produce low accuracy estimates when applied to environments different from their training environment, leading to increased user correction loads due to the need for manual correction of estimation results.
Innovation Solution
A teaching device that includes an acquisition unit, estimation unit, search unit, and selection unit to acquire input data, estimate results using a machine learning model, search for similar taught estimation results, and select a correction target from these results to reduce the correction load by using pre-existing similar data for correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the machine learning model is applied to a different environment, then the model can be used in new applications, but the estimation accuracy decreases
Solution Approach 1:
The system performs preliminary actions by automatically generating candidate correction results before user correction using similarity search and result synthesis. This preliminary correction reduces the burden on users and improves the efficiency of adapting the model to new environments while maintaining data quality for retraining
2Ease of operation
If the estimation result with low accuracy is used as correction target, then the correction process can proceed, but the user correction load increases
Solution Approach 1:
The system introduces an intermediary mechanism by generating candidate correction results through similarity search and synthesis before presenting to the user. This intermediary step provides the user with pre-processed correction options, significantly reducing the manual correction load and time required while maintaining ease of operation
Solution Approach 2:
The system performs preliminary correction work by automatically generating candidate results using similar cases from the teaching data. This preliminary action prepares multiple correction options in advance, allowing the user to select or refine rather than create corrections from scratch, thereby reducing correction load
Data Source
AI summary
According to an embodiment, a teaching device includes: an acquisition unit configured to acquire first input data; an estimation unit configured to estimate a first estimation result from the first input data, using a machine learning model; a search unit configured to search for a second taught estimation result taught for second input data, the second taught estimation result being associated with at least one of the second input data similar to the first input data, and a second estimation result similar to the first estimation result and estimated from the second input data, using the machine learning model; and a selection unit configured to select one selection candidate among a plurality of selection candidates including the first estimation result and the second taught estimation result, as a correction target estimation result to be used for correction of the first estimation result.