20240046151. METHOD AND ELECTRONIC DEVICE FOR AUTOMATED MACHINE LEARNING MODEL RETRAINING simplified abstract (SAMSUNG ELECTRONICS CO., LTD.)

From WikiPatents
Jump to navigation Jump to search

METHOD AND ELECTRONIC DEVICE FOR AUTOMATED MACHINE LEARNING MODEL RETRAINING

Organization Name

SAMSUNG ELECTRONICS CO., LTD.

Inventor(s)

Sukhdeep Singh of Bangalore (IN)

Vivek Sapru of Bangalore (IN)

Joseph Thaliath of Bangalore (IN)

Ganesh Kumar Thangavel of Bangalore (IN)

Ashish Jain of Bangalore (IN)

Seungil Yoon of SUWON-SI (KR)

Hoejoo Lee of Seoul (KR)

Hunje Yeon of Seoul (KR)

METHOD AND ELECTRONIC DEVICE FOR AUTOMATED MACHINE LEARNING MODEL RETRAINING - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240046151 titled 'METHOD AND ELECTRONIC DEVICE FOR AUTOMATED MACHINE LEARNING MODEL RETRAINING

Simplified Explanation

The abstract describes a system and method for automated machine learning (ML) model retraining by an electronic device. The system and method involve running a first ML model and a second ML model, predicting the accuracy degradation of the first ML model using the second ML model, determining if the predicted accuracy degradation meets a pre-defined threshold, and retraining the first ML model when the predicted accuracy degradation meets the threshold.

  • The system and method involve using multiple ML models to assess the accuracy degradation of a primary ML model.
  • The second ML model is used to predict the accuracy degradation of the first ML model.
  • The system determines if the predicted accuracy degradation meets a pre-defined threshold.
  • If the predicted accuracy degradation meets the threshold, the first ML model is retrained.

Potential applications of this technology:

  • Automated model retraining can be applied in various industries that rely on ML models, such as finance, healthcare, and e-commerce.
  • It can be used to improve the accuracy and performance of ML models over time without manual intervention.
  • This technology can be integrated into existing ML platforms and frameworks to enhance their capabilities.

Problems solved by this technology:

  • ML models can experience accuracy degradation over time due to changing data patterns or shifts in the underlying problem.
  • Manual retraining of ML models can be time-consuming and resource-intensive.
  • This technology automates the process of detecting accuracy degradation and initiating retraining, reducing the burden on human operators.

Benefits of this technology:

  • Improved accuracy: By automatically retraining ML models when accuracy degradation is predicted, the system ensures that models stay up-to-date and perform optimally.
  • Time and resource savings: Automated model retraining eliminates the need for manual intervention, saving time and resources for organizations.
  • Enhanced model performance: Continuous retraining allows ML models to adapt to changing data patterns, leading to improved performance and better decision-making.


Original Abstract Submitted

a system and/or method for automated ml model retraining by an electronic device. the system and/or method may include one or more of: running a first ml model and a second ml model, predicting an accuracy degradation of the first ml model using the second ml model, determining whether the predicted accuracy degradation meets a pre-defined threshold, and/or retraining the first ml model when the predicted accuracy degradation meets the pre-defined threshold.