18274262. CANDIDATE MACHINE LEARNING MODEL IDENTIFICATION AND SELECTION simplified abstract (Telefonaktiebolaget LM Ericsson (publ))

CANDIDATE MACHINE LEARNING MODEL IDENTIFICATION AND SELECTION

Organization Name

Telefonaktiebolaget LM Ericsson (publ)

Inventor(s)

Athanasios Karapentelakis of Solna (SE)

Alessandro Previti of Stockholm (SE)

Konstantinos Vandikas of Solna (SE)

Lackis Eleftheriadis of Valbo (SE)

Marin Orlic of Bromma (SE)

Marios Daoutis of Bromma (SE)

Maxim Teslenko of Sollentuna (SE)

Sai Hareesh Anamandra of Bangalore (IN)

CANDIDATE MACHINE LEARNING MODEL IDENTIFICATION AND SELECTION - A simplified explanation of the abstract

This abstract first appeared for US patent application 18274262 titled 'CANDIDATE MACHINE LEARNING MODEL IDENTIFICATION AND SELECTION

Simplified Explanation

The abstract describes a computer-implemented method for retrieving or executing machine learning models based on specified input and output features.

• The method involves receiving a request for a machine learning model with specific input data type and distribution.
• The method then matches the request with an identified machine learning model or combination of models that partially satisfies the request.
• A candidate model is selected based on its ability to produce the specified output feature.
• The selected model is further refined based on convergence criteria.

Potential Applications

This technology could be applied in various fields such as healthcare, finance, marketing, and more for optimizing machine learning model selection based on specific input and output requirements.

Problems Solved

This technology helps in efficiently identifying and selecting the most suitable machine learning models for a given task, reducing the time and effort required for manual model selection.

Benefits

The benefits of this technology include improved accuracy in model selection, faster deployment of machine learning solutions, and enhanced overall performance of machine learning applications.

Potential Commercial Applications

"Optimized Machine Learning Model Selection for Specific Input and Output Features"

This technology could be commercially applied in industries such as e-commerce, cybersecurity, and predictive maintenance for streamlining machine learning model selection processes.

Possible Prior Art

One possible prior art in this field is the use of automated machine learning (AutoML) tools that assist in automating the process of model selection and hyperparameter tuning.

Unanswered Questions

How does the method handle cases where multiple candidate models satisfy the input and output requirements equally well?

The abstract does not specify how the method resolves such scenarios where multiple candidate models meet the specified criteria.

What types of convergence criteria are used to select the final machine learning model?

The abstract mentions selecting a model based on convergence criteria, but it does not detail the specific criteria or algorithms used in the selection process.

Original Abstract Submitted

A computer-implemented method performed by a network node is provided. The method includes receiving a request for retrieving or executing a machine learning (ML) model or a combination of ML models. The request includes a first description of a specified output feature and specified input data type and distribution of input values for a ML model or combination of ML models. The method further includes obtaining an identification of a ML model, or a combination of ML models, having a second description that at least partially satisfies a match to the first description; identifying a candidate ML model, or combination of ML models, that produces the specified output feature of the first description based on a comparison of the first and second descriptions. The method further includes selecting a third description of the identified candidate ML model, or combination of ML models, based on a convergence.