Artificial Intelligence Model Factory

Patent metadata

Field	Value
Publication number	US-2025045624-A1
Application number	US-202318362726-A
Country	US
Kind code	A1
Filing date	Jul 31, 2023
Priority date	Jul 31, 2023
Publication date	Feb 6, 2025
Grant date	—

Abstract

An approach for generating an artificial intelligence system configurable for use with assets. In this approach, a model recipe is selected for generating the artificial intelligence system for use with assets. Recipe parameters specified in the model recipe are identified. A training dataset is created using the model recipe and input data. A set of artificial intelligence models is trained using the training dataset, the recipe parameters, and the model recipe. The training creates artifact models. The artifact models resulting from training are evaluated. The evaluation is used to select a set of the artifact models in the artifacts that form the artificial intelligence system that is configurable for use in assets.

First claim

What is claimed is:

1 . A computer implemented method for generating an artificial intelligence system, the computer implemented method comprising:
selecting, by a number of processor units, a model recipe for generating the artificial intelligence system for use with assets; identifying, by the number of processor units, recipe parameters specified in the model recipe; creating, by the number of processor units, a training dataset using the model recipe and input data; training, by the number of processor units, artificial intelligence models using the training dataset, the recipe parameters, and the model recipe to create artifact models; evaluating, by the number of processor units, artifact models resulting from training the artificial intelligence models to form an evaluation; and selecting, by the number of processor units, a set of the artifact models for the artificial intelligence system using the evaluation.

2 . The computer implemented method of claim 1 further comprising:
deploying, by the number of processor units, the artificial intelligence system to a set of target platforms.

3 . The computer implemented method of claim 2 , wherein deploying, by the number of processor units, the artificial intelligence system comprises:
identifying, by the number of processor units, a set of target platforms for the artificial intelligence system; creating, by the number of processor units, create a set of production artificial intelligence models to form artificial intelligence system to run on the target platforms using the model recipe and the set of artifact models; and deploying, by the number of processor units, the artificial intelligence system comprising the set of production artificial intelligence models to the set of target platforms using the model recipe.

4 . The computer implemented method of claim 2 further comprising:
monitoring, by the number of processor units, a number of performance metrics based on evaluation artifacts generated from training the artificial intelligence models; and
retraining, by the number of processor units, the artificial intelligence system based on the number of performance metrics.

5 . The computer implemented method of claim 1 , wherein training, by the number of processor units, the artificial intelligence models comprises:
identifying, by the number of processor units, resources for training the artificial intelligence models using the model recipe; creating, by the number of processor units, an execution cluster based on the resources identified; and training, by the number of processor units, the artificial intelligence models in the execution cluster using the training dataset and the model recipe.

6 . The computer implemented method of claim 1 , wherein training, by the number of processor units, the artificial intelligence models comprises:
creating, by the number of processor units, a project for creating the artifact models, wherein the project comprises multiple steps; and running, by the number of processor units, an experiment for the project that creates artifact models.

7 . The computer implemented method of claim 5 , wherein the model recipe is selected from a group comprising a generalized model recipe and customized model recipe.

8 . The computer implemented method of claim 1 , wherein the recipe parameters are selected from at least one of an application specific parameter, a learning parameter, and training level.

9 . The computer implemented method of claim 1 , wherein the artificial intelligence system comprises at least one of a single artificial intelligence model intelligence model for a single asset in the assets, multiple artificial intelligence models for each asset in the assets, a single artificial intelligence model for a group of assets in the assets, or multiple artificial intelligence models for multiple assets in the assets.

10 . A computer system for generating an artificial intelligence system, comprising:
one or more processors units; one or more computer readable storage devices; and computer program instructions, the computer program instructions being stored on the one or more computer readable storage devices for execution by the one or more processor units to perform one or more operations to:
select a model recipe for generating an artificial intelligence system for use with assets;
identify recipe parameters specified in the model recipe;
create a training dataset using the model recipe and input data;
train artificial intelligence models using the training dataset, the recipe parameters, and the model recipe to create artifact models;
evaluate the artifact models resulting from training the artificial intelligence models to form an evaluation; and
select a set of the artifact models for the artificial intelligence system using the evaluation.

11 . The computer system of claim 10 , the one or more processor units further executes the computer program instructions to:
deploy the artificial intelligence system to a set of target platforms.

12 . The computer system of claim 11 , wherein as part of deploying the artificial intelligence system, the one or more processor units further executes the computer program instructions to:
identify a set of target platforms for the artificial intelligence system; create a set of production artificial intelligence models to form artificial intelligence system to run on the target platforms using the model recipe and the set of artifact models; and deploy the artificial intelligence system comprising the set of production artificial intelligence models to the set of target platforms using the model recipe.

13 . The computer system of claim 11 , wherein the one or more processor units further executes the computer program instructions to:
monitor a number of performance metrics based on evaluation artifacts generated from training the artificial intelligence models; and retrain the artificial intelligence system based on the number of performance metrics.

14 . The computer system of claim 10 , wherein as part of training the artificial intelligence models, the one or more processor units further executes the computer program instructions to:
identify resources for training the artificial intelligence models using the model recipe; create an execution cluster based on the resources identified; and train the artificial intelligence models in the execution cluster using the training dataset and the model recipe.

15 . The computer system of claim 10 , wherein as part of training the artificial intelligence models, the one or more processor units further executes the computer program instructions to:
create a project for creating the artifact models, wherein the project comprises multiple steps; and run an experiment for the project that creates artifact models.

16 . The computer system of claim 14 , wherein the model recipe is selected from a group comprising a generalized model recipe and customized model recipe.

17 . The computer system of claim 10 , wherein the recipe parameters are selected from at least one of an application specific parameter, a learning parameter, and training level.

18 . The computer system of claim 10 , wherein the artificial intelligence system comprises at least one of a single artificial intelligence model intelligence model for a single asset in the assets, multiple artificial intelligence models for each asset in the assets, a single artificial intelligence model for a group of assets in the assets, or multiple arti

Assignees

• IBM

Inventors

• Patel Dhavalkumar C
• Sharma Vivek
• Bhamidipaty Anuradha
• Kalagnanam Jayant R
• Lin Shuxin
• Shah Dhruv
• Jayaraman Srideepika

Classifications

• G06N20/00 Primary

  Machine learning · CPC title