Information processing method, information processing device, and information processing program

Abstract

This information processing device: acquires a feature quantity indicating a feature of a measurement target measured by a sensor and predicts the state of the measurement target by entering the feature quantity into a machine learning model to improve the state prediction accuracy of the machine learning model; and acquires a plurality of design parameter modification methods for modifying design parameters of the sensor, determines the optimum design parameter modification method from among the plurality of design parameter modification methods on the basis of the feature quantity and the state prediction result, and outputs the determined optimum design parameter modification method.

Claims

1. A method of information processing in a computer, We obtain feature quantities that represent the characteristics of the object being measured by the sensor. By inputting the aforementioned features into a machine learning model, the state of the object to be measured is predicted. To improve the state prediction accuracy of the machine learning model and to obtain multiple design parameter modification methods for modifying the design parameters of the sensor, Based on the aforementioned feature quantities and the prediction results of the state, the optimal design parameter modification method is determined from among the multiple design parameter modification methods. Output the determined optimal design parameter modification method. Information processing methods.

2. In determining the optimal method for modifying the design parameters, Based on the aforementioned feature quantities and the predicted state, the amount of modification for each of the multiple design parameter modification methods is calculated. Based on the amount of modification, the optimal design parameter modification method is identified from among the multiple design parameter modification methods. The information processing method according to claim 1.

3. Furthermore, based on the predicted state and the correct state corresponding to the feature input to the machine learning model, it is determined whether the predicted state is the correct state or not. In determining the optimal method for modifying the design parameters, Furthermore, the design parameters for each of the multiple design parameter modification methods are calculated, Furthermore, the distance between the incorrect answer point in the feature space for the feature corresponding to the prediction result that was determined not to be in the correct state, and the correct answer point in the feature space for the feature corresponding to the prediction result that was determined to be in the correct state, is calculated as the prediction error. In calculating the amount of modification for the design parameters, the amount of modification in the design parameter space is calculated based on the calculated prediction error and the calculated design parameters. The information processing method according to claim 2.

4. In determining the optimal method for modifying the design parameters, Furthermore, a development cost coefficient is obtained, which is set for each of the multiple design parameter modification methods and according to the cost required for the development of the sensor. Furthermore, by multiplying the amount of modification for each of the multiple design parameter modification methods by the development cost coefficient for each of the multiple design parameter modification methods, the modification cost value for each of the multiple design parameter modification methods is calculated. In identifying the optimal design parameter modification method, the design parameter modification method that minimizes the calculated modification cost value is identified as the optimal design parameter modification method. The information processing method according to claim 2 or 3.

5. In determining the optimal design parameter modification method, the modification cost value for each of the multiple design parameter modification methods is further multiplied by a correction coefficient. In identifying the optimal design parameter modification method, the design parameter modification method that minimizes the modification cost value multiplied by the correction coefficient is identified as the optimal design parameter modification method. In predicting the state, the state of the object to be measured is predicted by inputting the feature quantities obtained from the sensor using the design parameters corrected by the identified optimal design parameter correction method into the machine learning model. Furthermore, based on the predicted state and the correct state corresponding to the feature input to the machine learning model, it is determined whether the predicted state is the correct state or not. In determining the optimal design parameter modification method, if it is determined that the predicted state is not the correct state, the correction coefficient is updated. The information processing method according to claim 4.

6. The design parameter is the mean value of the distribution of the feature quantities. The aforementioned method for modifying design parameters involves shifting the mean value of the distribution of the feature quantities. The information processing method according to any one of claims 1 to 3.

7. The design parameter is the standard deviation of the distribution of the feature quantities, The aforementioned method for modifying design parameters involves reducing the standard deviation of the distribution of the feature quantities. The information processing method according to any one of claims 1 to 3.

8. A feature quantity acquisition unit that acquires feature quantities that represent the characteristics of the object being measured by the sensor, A prediction unit that predicts the state of the object to be measured by inputting the aforementioned features into a machine learning model, A correction method acquisition unit that acquires multiple design parameter correction methods for improving the state prediction accuracy of the machine learning model and correcting the design parameters of the sensor, A modification method determination unit determines the optimal design parameter modification method from among the plurality of design parameter modification methods based on the aforementioned feature quantities and the prediction results of the state, An output unit that outputs the determined optimal design parameter modification method, An information processing device equipped with the following features.

9. We obtain feature quantities that represent the characteristics of the object being measured by the sensor. By inputting the aforementioned features into a machine learning model, the state of the object to be measured is predicted. To improve the state prediction accuracy of the machine learning model and to obtain multiple design parameter modification methods for modifying the design parameters of the sensor, Based on the aforementioned feature quantities and the prediction results of the state, the optimal design parameter modification method is determined from among the multiple design parameter modification methods. The computer is made to function in order to output the determined optimal method for modifying the design parameters. Information processing program.

Images