Verification and Validation for Trustworthy Scientific Machine Learning
This work addresses the problem of unreliable SciML models for researchers in scientific disciplines, but it is incremental as it builds on existing computational science guidelines.
The paper tackles the lack of trustworthiness in scientific machine learning (SciML) by proposing 16 recommendations to establish good practices for predictive SciML, focusing on verification and validation to improve modeling rigor.
Scientific machine learning (SciML) models are transforming many scientific disciplines. However, the development of good modeling practices to increase the trustworthiness of SciML has lagged behind its application, limiting its potential impact. The goal of this paper is to start a discussion on establishing consensus-based good practices for predictive SciML. We identify key challenges in applying existing computational science and engineering guidelines, such as verification and validation protocols, and provide recommendations to address these challenges. 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.