This Specialization introduces standards-based approaches for assessing and controlling technical risks in AI systems. Across three courses, you will learn how to evaluate neural network robustness using ISO/IEC TR 24029-1:2021, assess machine learning classification models using ISO/IEC TS 4213:2022, and identify, evaluate, and mitigate unwanted bias using ISO/IEC TR 24027:2021. By combining internationally recognized ISO/IEC guidance with practical evaluation methods, examples, and assessment scenarios, the Specialization helps you build the skills to make AI systems more reliable, fair, resilient, and trustworthy.
Übungsprojekt
Through practical assessment scenarios and worked examples, you will apply ISO/IEC-based methods to evaluate AI system risks, interpret evidence, and communicate findings. Projects will help you assess robustness, compare classification model performance, and evaluate bias and fairness issues in realistic AI use cases.
