麻豆村

Artificial intelligence is rapidly reshaping how scientists explore the universe 鈥� turning massive amounts of data into discoveries that were once out of reach. At 麻豆村, a new initiative will bring together experts in AI, statistics and astrophysics to accelerate that shift.聽

Supported by the Simons Foundation, the Keystone Astronomy & AI (KAAI) Visiting Fellows Program will accelerate the use of AI in cosmological and astronomical research through an international, mentored postdoctoral initiative.聽

KAAI Fellows will participate in a monthlong residency at the聽McWilliams Center for Cosmology & Astrophysics(opens in new window), where each visiting fellow is paired with two mentors 鈥� one in astrophysics and one in AI or statistics 鈥� to tackle high-impact problems at the intersection of astronomy and machine learning. Each residency culminates in a hands-on workshop that shares software, datasets and workflows with the broader community. The program aims to cultivate a globally connected cohort of researchers fluent in both astrophysics and modern machine learning while accelerating discovery in this data-rich scientific landscape.

The initiative also provides meaningful opportunities for Carnegie Mellon graduate students, who collaborate with visiting fellows, contribute to shared tools and workflows, and gain direct experience while applying AI to frontier problems in astrophysics.

鈥淎I is changing how we do science, and astronomy is where its impact will be felt first and fastest鈥� said聽Tiziana Di Matteo(opens in new window), director of the McWilliams Center and the primary investigator on this program.聽 鈥淲ith KAAI Fellows, we鈥檙e turning the McWilliams Center鈥檚 cross-disciplinary strength into a global training engine 鈥� bringing visiting scholars together with our machine-learning and astrophysics teams to develop methods that move the field and the way science is done.鈥�

The McWilliams Center fosters collaboration within Carnegie Mellon鈥檚聽Department of Physics(opens in new window), the, the聽Department of Statistics & Data Science (SDS)(opens in new window) and the and among partner institutions including the and the Department of Physics and Astronomy at the University of Pittsburgh.

A key to the program鈥檚 strength is the deep cross-disciplinary collaboration among researchers at the McWilliams Center, the Department of Machine Learning and the SDS and the聽STAtistical Methods for the Physical Sciences Research Center (STAMPS)(opens in new window), whose combined expertise forms the backbone of KAAI鈥檚 interdisciplinary model.

McWilliams researchers are developing the data science tools needed to process this immense stream of information into scientific breakthroughs that advance astrophysics and enable new technologies in fields like AI, imaging and data infrastructure on Earth.

The KAAI Fellows program will support six visiting fellows for a month each over the next three years. Applications will be open later this spring.聽

Visiting fellows will be selected for projects that integrate AI with theoretical and computational astrophysics, particularly in areas such as large-scale simulations, computational modeling and data-intensive analysis. By pairing each fellow with dual Carnegie Mellon mentors, the program fosters deep cross-disciplinary collaboration between domain scientists and AI experts.

, associate professor in Carnegie Mellon鈥檚, will serve as the program鈥檚 AI and machine learning director. A member of the McWilliams Center, P贸czos collaborates with other faculty, postdoctoral researchers and graduate students on shared code, data and computational tools.

鈥淚t is exciting to see how the newly developed machine-learning methods are transforming the way we approach science,鈥� P贸czos said. 鈥淚n astrophysics particularly, these tools are reshaping how we explore vast and complex datasets, enabling us to extract subtle signals, identify rare and interesting events, accelerate scientific simulations and test physical theories at unprecedented scale. By augmenting human intuition with data-driven discovery, machine learning has the potential to dramatically accelerate our understanding of the universe and uncover phenomena that would otherwise remain hidden.鈥�

Carnegie Mellon鈥檚 Machine Learning Department shares a long history of close collaboration with the McWilliams Center for Cosmology, combining expertise in machine learning, statistical inference and large-scale computation with deep domain knowledge in astrophysics. These sustained partnerships created impactful, collaborative research at the intersection of machine learning and cosmology, and continue to play a central role in advancing data-driven discovery in the physical sciences.

Fellows will leave the program with demonstrated experience applying trustworthy AI to frontier astrophysics and with durable connections that extend beyond astronomy.

A core component of the fellowship is knowledge dissemination. At the end of each visit, each KAAI Fellow will co-organize a weeklong, hands-on workshop showcasing cutting-edge AI methods for astronomy. These workshops will help accelerate the adoption of new tools across the international research community, ensuring the advanced approaches spread well beyond individual projects or institutions. Designed for maximum impact, they also will cultivate a global network of researchers skilled in applying state-of-the-art techniques to fundamental questions about the universe.

鈥淲e鈥檙e working to develop a global community of international experts in subfields related to AI and astronomy,鈥� Di Matteo said. 鈥淪upported by Simons, the workshops will bring together experts from machine learning and astronomy to drive the field forward.鈥�