When working on the discovery of new materials, most machine-learning models today only consider a few specific types of data or variables. But now, a team, including researchers from MIT DMSE and NSE, has developed a method for optimizing materials recipes and planning experiments that incorporates information from diverse sources like insights from the literature, chemical compositions, microstructural images, and more. https://news.mit.edu/2025/ai-system-learns-many-types-scientific-informati