When: Wednesday, February 26, 2025 from 2pm – 3pm
Where: Amy Gutmann Hall, Room 414
Title: “Closing the Gap Between Scientific Foundation Models and Real-World Applications”
Abstract: In the era of LLM models, one gets notoriously confronted with the question of where we stand with the applicability of large-scale deep learning models within scientific or engineering domains. The talk is motivated by recent triumphs in weather and climate modeling, and discusses potentials, breakthroughs, and remaining challenges in fluid dynamics and related engineering fields. Concretely, we showcase recent work in scaling neural networks to model multi-physics phenomena and computational fluid dynamics as used in automotive engineering. Finally, we outline challenges and potential solutions when it comes to scalability beyond traditional numerical schemes and discuss the respective impact on industry and scientific environments.
Bio: Johannes Brandstetter is leading a group on “AI for data-driven simulations” at the Institute for Machine Learning at the Johannes Kepler University (JKU) Linz. Additionally, he is a Chief Researcher at NXAI – their new European AI hub in Linz (Austria). He obtained his PhD working on Higgs boson physics at the CMS experiment at CERN, and since then has worked in Deep Learning, AI4Science, and neighboring interdisciplinary fields. His interests comprise data-driven simulations, Geometric Deep Learning, and possible extensions to engineering and scientific applications.
We’re excited to announce Data Points, a new research blog showcasing the work of Penn’s data science community!
Data Points aims to make data science research more accessible through thoughtful reviews, high-level walkthroughs, and interactive, explorable explanations. The blog series provides a look at the research and researchers currently shaping data science at Penn.
Check out our first post written by AI x Science fellow, Kieran Murphy, here.
Interested in contributing? Learn more in our How to Contribute guide. We’re excited to collaborate and help bring your ideas to life!
We’re thrilled to share that the Data-Driven Discovery Initiative was recently highlighted in an article by Omnia, the alumni magazine for the University of Pennsylvania School of Arts and Sciences. The piece, titled “The Missing Link: Connecting Staff and Data Scientists Across the University,” shines a spotlight on our work with IDEAS (SEAS) and IBI (PSOM) to support Penn’s data science community.
Read the full story here to learn more about how we’re connecting data scientists Penn-wide!
The Center for Innovation in Data Engineering and Science (IDEAS) and the Data-Driven Discovery Initiative (DDDI) welcome the inaugural cohort of the Penn AI x Science Postdoctoral Fellows Program! This unique fellowship supports postdoctoral researchers working at the intersection of science, machine learning, and artificial intelligence across the School of Engineering and Applied Science (SEAS) and the School of Arts and Sciences (SAS).
The program’s mission is to foster innovation and excellence by promoting interdisciplinary collaboration among fellows and faculty from across Penn’s schools. Fellows benefit from sustained peer interactions, broad exposure to world-class faculty, and opportunities for cross-school mentorship. The new AI x Science fellows will join a community of exceptional Data-Driven Discovery postdoctoral fellows from across the natural and social sciences.
This year, we are excited to welcome Noemi Aepli (linguistics diversity & NLP), Yahav Bechavod (reliable & fair ML), Sourav Dey (chemistry & ML), Nicolò Dal Fabbro (distributed multi-agent AI), Marcelo Guzmán (physics-based learning), Xinquan Huang (neural PDEs & digital twins), Hancheng Min (dynamical systems & ML), Kieran Murphy (information localization), Melanie Segado (movement-based biomarkers), Brynn Sherman (cognition & memory), and Yan Sun (trustworthy ML & model calibration) as the first cohort of the Penn AI x Science Postdoctoral Fellows Program. Congratulations to all our new fellows!
Applications for the DDDI Postdoctoral Fellowship are now open for the 2025/2026 cycle!
The DDDI Postdoctoral Fellowship is an opportunity open to all postdocs in the School of Arts & Sciences who use data science as a significant part of their research methodology. We encourage postdocs across the natural and social sciences to apply.
Accepted applicants will join a group of data-driven postdoctoral fellows who regularly meet together as well as with faculty and visitors to discuss their research and exchange ideas. Fellows will receive a research stipend (up to $5,000 total) for one year to support their research efforts.
The application deadline is January 15th, 2025. Information on how to apply can be found here.
For more details, please visit our information for faculty page, or contact crtwomey@sas.upenn.edu.
