In the scientific report titled “Verification and validation for trustworthy scientific machine learning,” GW Engineering’s Lorena Barba, a professor of mechanical and aerospace engineering, contributed to research on advancing trustworthy science machine learning (ScIML) models. The study, published in the highly ranked IOP journal Machine Learning Science and Technology, explores challenges in applying existing verification and validation protocols to SciML models and recommendations for addressing them.
Here is an excerpt from the study abstract: “Our discussion focuses on predictive SciML, which uses machine learning models to learn, improve, and accelerate numerical simulations of physical systems. While centered on predictive applications, our 16 recommendations aim to help researchers conduct and document their modeling processes rigorously across all SciML domains.”
Read the full study in Machine Learning: Science and Technology.