KGML Lab

Welcome to the Knowledge-guided Machine Learning (KGML) lab led by Prof. Anuj Karpatne at Virginia Tech in the Department of Computer Science and the Sanghani Center for AI and Data Analytics.

A central focus of our lab is to advance the emerging field of KGML where scientific knowledge is deeply integrated in the design and training of ML models to produce scientifically grounded, explainable, and generalizable results, going beyond black-box (data-only) applications of AI/ML in science. Through our inter-disciplinary collaborations with researchers from diverse institutions and disciplinary backgrounds, we aspire to contribute on two fundamental fronts: (1) Advance the foundations of AI/ML by incorporating diverse forms of scientific knowledge in AI/ML frameworks including partial differential equations (PDEs), symbolic rules, ontologies, and mechanistic models, and (2) Deliver real-world impacts to scientific applications of high societal relevance including aquatic sciences, organismal biology, virology, mechanobiology, fluid dynamics, geophysics, quantum mechanics, and electromagnetism. We are grateful to NSF for their generous support for our research projects. Check out our Projects, Publications, and Team pages to learn more about us and our work.

To learn more about the field of KGML, see the KGML book and a recent perspective article summarizing current trends and future prospects in this emerging field.