17978511. MODEL LEARNING SYSTEM AND MODEL LEARNING DEVICE simplified abstract (TOYOTA JIDOSHA KABUSHIKI KAISHA)

MODEL LEARNING SYSTEM AND MODEL LEARNING DEVICE

Organization Name

TOYOTA JIDOSHA KABUSHIKI KAISHA

Inventor(s)

Daiki Yokoyama of Gotemba-shi Shizuoka-ken (JP)

Tomohiro Kaneko of Mishima-shi Shizuoka-ken (JP)

MODEL LEARNING SYSTEM AND MODEL LEARNING DEVICE - A simplified explanation of the abstract

This abstract first appeared for US patent application 17978511 titled 'MODEL LEARNING SYSTEM AND MODEL LEARNING DEVICE

Simplified Explanation

Abstract: A model learning system is described that includes a device for learning models, a data acquisition device for collecting data to create a training data set, and a model learning device that automatically learns the model using the training data set. The model learning device is designed to exclude a learning model from being automatically learned if its accuracy decreases or does not improve compared to before learning.

Patent/Innovation Explanation:

• The system includes a device for learning models.
• A data acquisition device is used to collect data for creating a training data set.
• A model learning device automatically learns the model using the training data set.
• If the accuracy of the model decreases or does not improve after learning, the model learning device excludes it from being automatically learned.

Potential Applications:

• This technology can be applied in various fields that require model learning, such as machine learning, artificial intelligence, and data analysis.
• It can be used in industries like finance, healthcare, marketing, and manufacturing to improve decision-making processes and optimize operations.

Problems Solved:

• The system addresses the issue of learning models that do not perform well or provide accurate results.
• It solves the problem of wasting resources on learning models that do not improve or decrease in accuracy after training.

Benefits of this Technology:

• The system ensures that only effective and accurate models are learned and utilized.
• It saves time and resources by excluding models that do not provide improved accuracy.
• It improves the overall performance and reliability of the learning system.

Original Abstract Submitted

A model learning system includes a device in which a learning model is used, a data acquisition device that acquires data used for creating a training data set for the learning model, and a model learning device that automatically learns the learning model using the training data set for the learning model created using the data. The model learning device is configured to, when a model accuracy after automatically learning the learning model decreases or does not increase compared to before learning, exclude the learning model from a target to be automatically learned.