T/SAIAS 001-2024 Active Group standards

T/SAIAS 001-2024 Neural network classification model robustness testing method

Publish Date: 2024-08-31 Implement Date: 2024-09-01 For services related to genuine standard inquiry, procurement, translation, and other related services in China, please Contact Us

Price: $ 20.00

Out of stock

Basic Information

Standard Code: T/SAIAS 001-2024

Standard Type: Group standards

Standard Status: Active

is_force_gb: no

CCS Name: -

ICS Name: Integrated Information Technology (IT)

Publish Date: 2024-08-31

Implement Date: 2024-09-01

Scope

Scope: This document proposes factors to be considered for the robustness of neural network classification models and provides a method for testing the robustness of neural network classification models. This document is applicable to the robustness testing and evaluation system for neural network classification models. Main technical content: This standard project focuses on the robustness dimension and develops standards related to the basic requirements and testing methods for the robustness of neural network classification models. Improving the robustness of models poses two-level requirements: first, robustness enhancement technologies; second, robustness measurement. The development of robustness measurement standards will help establish a unified evaluation system and framework, set new practical robustness standards for industry applications of models, and thereby enhance the overall trustworthiness of the AI industry, playing a positive role in promoting the development of trustworthy AI. This standard specifies the basic requirements for the robustness of neural network classification models and provides a method for testing the robustness of neural network classification models.