We are excited to welcome Prof. Michael Mahoney from UC Berkeley for a distinguished lecture on “Foundational Methods for Foundation Models for Scientific Machine Learning”

In this talk, Prof. Mahoney will explore how foundation models, like those powering ChatGPT and cutting-edge computer vision, can be adapted and advanced to serve the needs of scientific machine learning (SciML). He will present groundbreaking work on applying the "pre-train and fine-tune" paradigm to large-scale scientific data and discuss key challenges and failure modes in integrating machine learning with traditional scientific computing. The talk will also cover novel algorithmic approaches designed to overcome these hurdles at scale.

About the speaker:

Michael W. Mahoney is a Professor in the Department of Statistics at the University of California, Berkeley, and an Amazon Scholar. He also leads the Machine Learning and Analytics Group at the Lawrence Berkeley National Lab. His research spans large-scale machine learning, randomized algorithms, scientific computing, and data applications across multiple domains. 

-Date: March 26, 2025

-Time: 14:00 CET

-Location: Lecture Hall W201, Professor-Huber-Platz 2, LMU Munich

For further details, click here.

Everyone is welcome! No registration required.