Verification, Validation, and Uncertainty Quantification of Machine Learning Models

Machine-learning models have generated a lot of interest in thermal hydraulics and are starting to penetrate the practice of engineering and design. However, the methods on verification, validation and uncertainty quantification are not routinely applied and the methodology is still in development. The panelists will discuss their experience with ML-based model applications and development and the challenges of V&V and UQ approaches towards ML.