AI Model Validation Workshop

Service Description

Topics covered: • Understand the specifics of validating Artificial Intelligence models • Identify the differences between validating a conventional computerized system and validating an AI model • Master the AI model lifecycle: design, training, testing, deployment and monitoring • Define measurable requirements: performance, robustness, explainability and security • Ensure the quality, traceability and integrity of training and test data • Implement a risk-based approach suited to AI systems • Manage algorithmic bias and fairness issues • Structure documentation and traceability (datasets, versions, changes) By the end of this training, your teams will be able to: • Understand the regulatory and quality challenges of AI models • Define a validation strategy proportionate to the risk level • Identify critical points related to data and algorithms • Set up governance suited to AI projects • Maintain the model's compliance and performance over time