ISO/IEC 24029 Artificial Intelligence — Assessment of the robustness of neural networks

ISO/IEC 24029 is an international standard that provides guidelines for designing, developing, implementing, and using Artificial Intelligence (AI) systems. It covers various topics concerning the robustness of neural networks, including data management, evaluation and validation, and AI deployment and operation. The standard is divided into two parts: Part 1 provides general guidelines and principles for the design, development, implementation, and use of AI systems, while Part 2 provides additional guidelines and recommendations for the design, development, implementation, and use of AI systems in specific scenarios and environments. The standard is intended to be used as a framework for organizations of all sizes and across all industries that are involved in the development or use of AI systems, and it provides guidance on how to design, develop, implement, and use AI systems that are safe, reliable, and trustworthy.


Day 1 . The training covers the following topics

    • General processes for assessment of machine learning classification performance
    • Statistical tests and measures for performance assessment
    • Overview of technical methods for robustness assessment of neural networks
    • Robustness assessment model
    • Formal methods for robustness assessment
    • Data perturbation

Day 2 . 2h exam covering the course topics

Please find our additional resources below: