What is scientific machine learning?

Scientific Machine Learning (SciML) is an interdisciplinary approach that integrates traditional scientific modeling with state-of-the-art machine learning techniques. By bridging the gap between mechanistic models and data-driven machine learning methods, SciML provides a robust framework for understanding, predicting, and controlling complex systems.

At the vanguard of Scientific Machine Learning, our lab endeavors to create a harmonious synthesis between time-honored scientific modeling and the transformative capabilities of modern artificial intelligence (AI). While each paradigm has individually revolutionized their domains, their combined potential remains a frontier we’re passionate about exploring.

Our mission is twofold: to establish a foundational framework that seamlessly integrates mechanistic models with machine learning, and to translate this theoretical foundation into tangible ‘digital twins’. These digital entities are more than mere simulations; they’re dynamic reflections of real-world systems.

Grounded in disciplines spanning from nonparametric statistics to dynamical systems, causality and Physics-Informed Machine Learning, our research introduces novel methodologies tailored for diverse sectors. These range from psychotherapy and agriculture to  queuing theory, infectious diseases and climate dynamics. 

In essence, our lab is not just about combining two scientific paradigms; it’s about shaping a future where traditional scientific wisdom coexists and thrives alongside AI-driven innovations, fostering a brighter, more informed tomorrow.