|October 5, 2021
|Download as iCal:
Dr. Sven Krippendorf
Arnold Sommerfeld Center for Theoretical Physics, München
Dienstag, 5. Oktober 2021, 14:00 Uhr
“Theoretical Physicists' Biases Meet Machine Learning”
Many recent successes in machine learning (ML) resemble the success story in theoretical particle physics of utilising symmetries as organising principle. I discuss an introductory example where this procedure applied in ML leads to new insights to PDEs in mathematical physics, more precisely for the study of Calabi-Yau metrics. Then I discuss methods on how to identify symmetries of a system without requiring knowledge about such symmetries, including also how to find a Lax pair/connection associated with integrable systems. If time permits, I discuss how latent representations in neural networks can offer a close resemblance of variables used in dual descriptions established analytically in physical systems.