# US 2026/0073282 A1 · Service feasibility assessment and monitoring for initial ai model handling

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0073282 A1
- Document ID: US20260073282A1
- Status: Published application
- Kind: A1
- Application number: 18/829,777
- Filed: 2024-09-10
- Published: 2026-03-12
- Earliest claimed: 2024-09-10
- Assignees: Huawei Technologies Co., Ltd.
- Applicants: Huawei Technologies Co., Ltd.
- Inventors: Hesham Gamal Aly Mohamed MOUSSA; Mehdi Arashmid AKHAVAIN MOHAMMADI
- Web page: https://patentgrep.com/patent/US20260073282A1

## CPC

- G06N 20/00: Machine learning
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Embodiments of the present application provide a method, apparatus, and computer readable medium for evaluating service feasibility of AI models training on a global dataset. At least one AI model, data requirements for the AI model, and a target training accuracy performance indicator for the AI model when trained are received from a user. Relevant data is determined from a global dataset, the relevant data is based on the received data requirements. A training accuracy estimation module is used to calculate a model score for the AI model based on the relevant data, the model score representing an estimated training accuracy to which the AI model can be trained. Based on the model score, it is determined whether the target training accuracy performance indicator can be satisfied. When the target training accuracy performance indicator can be satisfied, the at least one AI model is communicated to a model training environment.

## Claims

### Claim 1

A method comprising:
receiving at least one artificial intelligence (AI) model, data requirements for the at least one AI model, and a target training accuracy performance indicator for the at least one AI model when trained;
determining relevant data from a global dataset, the relevant data based on the received data requirements;
using a training accuracy estimation module to calculate a model score for the at least one AI model based on the relevant data, the model score representing an estimated training accuracy to which the at least one AI model can be trained;
determining, based on the model score, whether the target training accuracy performance indicator can be satisfied; and
communicating the at least one AI model to a model training environment when the target training accuracy performance indicator can be satisfied.

### Claim 2 (depends on 1)

The method of claim 1 further comprising, when the target training accuracy performance indicator cannot be satisfied:
analysing information from the model training environment; and
based on the analysing, generating suggested modifications to the at least one AI model to improve the likelihood that the target training accuracy performance indicator can be satisfied.

### Claim 3 (depends on 1)

The method of claim 1 further comprising, when the target training accuracy performance indicator cannot be satisfied:
waiting an amount of time;
retrieving updated relevant data from the global dataset;
using the training accuracy estimation module to re-process the at least one AI model based on the updated relevant data to calculate a new model score;
determining, based on the new model score, whether the target training accuracy performance indicator can be satisfied; and
communicating the at least one AI models to the model training environment when the target training accuracy performance indicator can be satisfied.

### Claim 4 (depends on 1)

The method of claim 1 further comprising, when the target training accuracy performance indicator cannot be satisfied:
identifying at least one parameter likely to improve the accuracy to which the at least one AI model can be trained;
in response to the identified at least one parameter being updated, retrieving the relevant data, wherein the relevant data includes the updated parameter;
using the training accuracy estimation module to reprocess the at least one AI model based on the relevant data with the at least one updated parameter to calculate a new model score;
determining based on the new model score, whether the target training accuracy performance indicator can be satisfied; and
communicating the at least one AI model to the model training environment when the target training accuracy performance indicator can be satisfied.

### Claim 5 (depends on 1)

The method of claim 1 further comprising, when the target training accuracy performance indicator can be satisfied:
monitoring at least one parameter likely to affect the performance of the at least one AI model; and
in response to determining the at least one parameter affects performance of the at least one AI model beyond a predefined performance threshold, re-training the at least one AI model.

### Claim 6 (depends on 5)

The method of claim 5, wherein the at least one parameter is monitored automatically or updated in response to a user inquiry to determine if it affects performance of the at least one AI model beyond the predefined performance threshold.

### Claim 7 (depends on 1)

The method of claim 1, wherein the relevant data comprises actual data from the global dataset or a representation of the actual data from the global dataset.

### Claim 8 (depends on 1)

The method of claim 1 further comprising, when the target training accuracy performance indicator can be satisfied:
determining whether a model engine queue is experiencing at least one of storage or processing constraints, the model engine queue configured to queue AI models before entry to the model training environment; and
regulating communication of the at least one AI model to the model engine queue if the model engine queue is experiencing the at least one of storage or processing constraints.

### Claim 9 (depends on 1)

The method of claim 1 further comprising prioritizing access to the model training environment for different ones of the at least one AI models.

### Claim 10 (depends on 1)

The method of claim 1, wherein the training accuracy estimator comprises a zero-cost proxy and the method further comprises providing the at least one AI model to the zero-cost proxy for calculating the model score using the relevant data from the global dataset.

### Claim 11

Apparatus comprising:
one or more processors; and
a non-transitory computer readable medium having stored thereon instructions which, when executed by the one or more processors, cause the apparatus to:
receive at least one artificial intelligence (AI) model, data requirements for the at least one AI model, and a target training accuracy performance indicator for the at least one AI model when trained;
determine relevant data from a global dataset, the relevant data based on the received data requirements;
use a training accuracy estimation module to calculate a model score for the at least one AI model based on the relevant data, the model score representing an estimated training accuracy to which the at least one AI model can be trained;
determine, based on the model score, whether the target training performance indicator can be satisfied; and
communicate the at least one AI model to a model training environment when the target training accuracy performance indicator can be satisfied.

### Claim 12 (depends on 11)

The apparatus of claim 11 further comprising instructions that, when the target training accuracy performance indicator cannot be satisfied, cause the apparatus to:
analyse information from the model training environment; and
generate suggested modifications to the AI model to improve the likelihood that the target training accuracy performance indicator can be satisfied.

### Claim 13 (depends on 11)

The apparatus of claim 11 further comprising instructions that, when the target training accuracy performance indicator cannot be satisfied, cause the apparatus to:
wait an amount of time;
retrieve updated relevant data from the global dataset;
use the training accuracy estimation module to re-process the at least one AI model based on the updated relevant data to calculate a new model score;
determine whether the new model score satisfies the target training accuracy performance indicator; and
communicate the at least one AI model to the model training environment when the new model score satisfies the target key performance indicator.

### Claim 14 (depends on 11)

The apparatus of claim 11 further comprising instructions that, when the target training accuracy performance indicator cannot be satisfied, cause the apparatus to:
identify at least one parameter likely to improve the accuracy to which the at least one AI model can be trained;
in response to the identified parameter being updated, retrieve the relevant data, wherein the relevant data includes the updated parameter;
use the training accuracy estimation module to-reprocess the at least one AI model based on the relevant data with the at least one updated parameter to calculate a new model score;
determine, based on the new model score, whether the target training accuracy performance indicator can be satisfied; and
communicate the at least one AI model to the model training environment when the target training accuracy performance indicator can be satisfied.

### Claim 15 (depends on 11)

The apparatus of claim 11 further comprising instructions that, when the target training accuracy performance indicator can be satisfied, cause the apparatus to:
monitor at least one parameter likely to affect the performance of the at least one AI model; and
in response to determining that the at least one parameter affects performance of the at least one AI model beyond a predefined performance threshold, re-train the at least one AI model.

### Claim 16 (depends on 15)

The apparatus of claim 15, wherein the at least one parameter is monitored automatically or updated in response to a user inquiry to determine if it affects operation of the at least one AI model beyond the predefined operational threshold.

### Claim 17 (depends on 11)

The apparatus of claim 11 further comprising instructions that, when the target training accuracy performance indicator can be satisfied, cause the apparatus:
determine that a model engine queue is experiencing at least one of storage or processing constraints, the model engine queue configured to queue AI models before entry to the model training environment; and
regulate communication of the at least one AI model to the engine queue if the model engine queue is experiencing the at least one of storage or processing constraints.

### Claim 18 (depends on 11)

The apparatus of claim 11 further comprising prioritizing access to the model training environment for different ones of the at least one AI model.

### Claim 19 (depends on 11)

The apparatus of claim 11, wherein the training accuracy estimator comprises a zero-cost proxy and the method further comprises providing the at least one AI model to the zero-cost proxy for calculating the model score using the relevant data from the global dataset.

### Claim 20 (depends on 11)

The apparatus of claim 11, wherein the apparatus is a component in a service-based architecture for deploying AI-based components within a network.