International business machines corporation (20240232689). INTELLIGENT DEVICE DATA FILTER FOR MACHINE LEARNING simplified abstract

INTELLIGENT DEVICE DATA FILTER FOR MACHINE LEARNING

Organization Name

international business machines corporation

Inventor(s)

Janice Yang of Cambridge MA (US)

Kate Hogan of St. Louis MO (US)

Arnav Aggarwal of Los Altos CA (US)

Ethan Weaver of Morgantown WV (US)

John S. Werner of Fishkill NY (US)

INTELLIGENT DEVICE DATA FILTER FOR MACHINE LEARNING - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240232689 titled 'INTELLIGENT DEVICE DATA FILTER FOR MACHINE LEARNING

Simplified Explanation

The patent application describes a system where a processor at an edge device running a local instance of a machine learning model receives inference data. The processor then determines the probability for each data point to decide if it should be sent back to the cloud environment for retraining the model based on a set threshold.

• The processor at the edge device evaluates inference data to decide if it should be sent back to the cloud for retraining the machine learning model.
• The system aims to improve the accuracy and efficiency of the machine learning model by selectively sending data points back to the cloud for retraining based on set criteria.
• By running data points through filters and analyzing probabilities, the system optimizes the retraining process of the machine learning model.
• This approach helps in enhancing the performance of the machine learning model deployed at the edge device by continuously updating it with relevant data points.
• The system ensures that only data points meeting specific criteria are sent back to the cloud environment for retraining, reducing unnecessary data transfer and improving overall model performance.

Key Features and Innovation

• Processor at edge device evaluates inference data for retraining the machine learning model.
• Selective sending of data points back to the cloud based on set criteria.
• Optimization of retraining process by analyzing probabilities of data points.
• Continuous improvement of model performance by updating with relevant data.
• Efficient utilization of resources by sending only relevant data points for retraining.

Potential Applications

The technology can be applied in various fields such as:

• Internet of Things (IoT) devices
• Autonomous vehicles
• Healthcare for personalized treatment
• Predictive maintenance in manufacturing
• Fraud detection in financial services

Problems Solved

• Efficient retraining of machine learning models at edge devices.
• Reduction of unnecessary data transfer to the cloud.
• Improved accuracy and performance of machine learning models.
• Optimization of resources for retraining processes.
• Real-time updating of models with relevant data points.

Benefits

• Enhanced accuracy and efficiency of machine learning models.
• Cost-effective retraining processes.
• Real-time model updates for better performance.
• Reduction in data transfer costs.
• Improved decision-making based on updated models.

Commercial Applications

Selective Data Retraining System for Edge Devices

This technology can be utilized by companies developing edge computing solutions for various industries. By offering a system that optimizes the retraining process of machine learning models at edge devices, businesses can improve the performance and accuracy of their applications, leading to better customer satisfaction and competitive advantage in the market.

Prior Art

There may be prior art related to selective data retraining systems for machine learning models at edge devices. Researchers and developers can explore academic publications, patent databases, and industry reports to find similar technologies or approaches in this field.

Frequently Updated Research

Researchers are continuously exploring ways to enhance the efficiency and accuracy of machine learning models at edge devices. Stay updated on recent studies, conferences, and publications related to selective data retraining systems for edge computing to gain insights into the latest advancements in this area.

Questions about Selective Data Retraining System for Edge Devices

How does the system determine the probability for each data point to be sent back to the cloud for retraining?

The system runs data points through filters to analyze probabilities based on set criteria before deciding to send them back to the cloud for retraining.

What are the potential applications of this technology beyond edge devices?

This technology can also be applied in various fields such as IoT devices, autonomous vehicles, healthcare, manufacturing, and financial services for improving model performance and decision-making processes.

Original Abstract Submitted

in an approach, a processor receives, at an edge device running a local instance of a machine learning (ml) model, a set of inference data comprising a plurality of datapoints, wherein the local instance of the ml model is a deployed version of the ml model running in a cloud environment, and wherein the ml model was trained in the cloud environment and then deployed to the edge device. a processor runs the plurality of datapoints through one or more filters to determine a probability for each datapoint of whether a respective datapoint should be sent back to the cloud environment and used for retraining the ml model. a processor determines, for each datapoint, whether the probability for the respective datapoint meets a send back threshold that is required to be met before the respective datapoint is sent back to the cloud environment.