18275310. Method and Apparatus for Selecting Machine Learning Model for Execution in a Resource Constraint Environment simplified abstract (Telefonaktiebolaget LM Ericsson (publ))

Method and Apparatus for Selecting Machine Learning Model for Execution in a Resource Constraint Environment

Organization Name

Telefonaktiebolaget LM Ericsson (publ)

Inventor(s)

Andreas Johnsson of Uppsala (SE)

Rerngvit Yanggratoke of Järfälla (SE)

Method and Apparatus for Selecting Machine Learning Model for Execution in a Resource Constraint Environment - A simplified explanation of the abstract

This abstract first appeared for US patent application 18275310 titled 'Method and Apparatus for Selecting Machine Learning Model for Execution in a Resource Constraint Environment

Simplified Explanation

Embodiments herein disclose a method for selecting a machine learning model to be deployed in an execution environment with resource constraints. The method involves receiving a request for a machine learning model solving a task T using a feature set F, retrieving a set of machine learning models from a model store that solves the task T using at least a subset of features F, calculating the complexity of each model, and determining at least one suitable machine learning model to be deployed based on the calculated complexity and the resource constraints of the execution environment.

• The method involves receiving a request for a machine learning model solving a specific task using a set of features.
• The method retrieves a set of machine learning models from a model store that can solve the task using a subset of the features.
• The complexity of each model is calculated to determine the most suitable model to be deployed based on resource constraints.

Potential Applications

This technology could be applied in various fields such as healthcare, finance, and manufacturing where resource-constrained environments require efficient deployment of machine learning models.

Problems Solved

This technology solves the problem of selecting the most suitable machine learning model for deployment in resource-constrained execution environments, optimizing performance and resource utilization.

Benefits

The benefits of this technology include improved efficiency in deploying machine learning models, better utilization of resources, and enhanced performance in resource-constrained environments.

Potential Commercial Applications

Potential commercial applications of this technology include automated model selection in edge computing, IoT devices, and cloud computing environments, where resource constraints are a concern.

Possible Prior Art

One possible prior art in this field is the use of model selection algorithms based on performance metrics and resource constraints to optimize machine learning model deployment in various environments.

How does this method handle dynamic changes in resource constraints during model deployment?

The method does not address how it adapts to dynamic changes in resource constraints during the deployment of machine learning models.

What impact does the complexity calculation have on the overall performance of the selected machine learning model?

The method does not discuss the potential impact of the complexity calculation on the performance of the selected machine learning model.

Original Abstract Submitted

Embodiments herein disclose a method for selecting a machine learning model to be deployed in an execution environment having resource constraints. The method comprises receiving, by an apparatus, a request for a machine learning model solving a task T using a feature set F. Further, the method includes retrieving, from a model store, a first set of machine learning models that solves the task T using at least a subset of features F. The complexity of each machine learning model in the first set of machine learning models is calculated. The method includes determining, from the first set of machine learning models, at least one suitable machine learning model to be deployed, wherein the determining is based on the calculated complexity and the resource constraints of the execution environment.