Melanie Weber

Melanie Weber is an Assistant Professor of Applied Mathematics and Computer Science at Harvard University, where she leads the Geometric Machine Learning Group. Her research studies geometric structure in data and models and how to leverage such information for the design of new, efficient Machine Learning algorithms with provable guarantees. In 2021-2022, she was a Hooke Research Fellow at the Mathematical Institute in Oxford. Previously, she received her PhD from Princeton University (2021) and undergraduate degrees in Mathematics and Physics from the University of Leipzig, Germany (2016). She is a recipient of the IMA Leslie Fox Prize in Numerical Analysis, an Alfred P. Sloan Research Fellowship in Mathematics, an Aramont Fellowship for Emerging Science Research, and an Airforce Young Investigator Award.

AI2050 Project

Artificial intelligence is revolutionizing science, with foundation models driving breakthroughs in climate solutions, disease treatment, and materials design. However, these models require vast data and computing resources, which are often unavailable in the sciences and raise sustainability concerns. Encoding geometric structure, such as symmetries arising from laws of physics, can boost efficiency by narrowing the model’s focus to physically plausible conditions. Current geometric models excel in specific tasks but lack broad applicability to complex scientific problems. Weber aims to develop geometry-informed models that combine the strengths of geometric and general-purpose foundation models, building resource-efficient AI tools for diverse scientific challenges.