MULYADI / UNSPLASH

News

LOREDANA GASTALDO

ERC Advanced Grant for Loredana Gastaldo

The ERC is funding the project “Electron Capture in Ho-163 – Large Experiment” (ECHo-LE)   more ...

AIM Connects Students and Research Groups

Around 500 students explored research opportunities at the AIM Summer Edition.   more ...
NSF–DOE Vera C. Rubin Observatory

Vera C. Rubin Observatory Captures First Images

ARI appeared in the Tagesschau, as the first images from the Vera C. Rubin Observatory in Chile were presented.   more ...
FABIENNE GANTENBEIN

NTMxISOQUANT SciArt Residency Launches with Award-Winning Playwright

ISOQUANT and the Nationaltheater Mannheim launch a residency that bridges quantum physics and the performing arts.   more ...
SANDRA KLEVANSKY

Teaching Awards Announced

Four instructors were recognized for particularly effective teaching in the winter semester 2024/25.   more ...
ALESSA KLIOBA

Felix Röper Defends His Title at the Heidelberg Integration Bee

In front of nearly 300 spectators, students tackled challenging integrals —including those by Fields Medalists — at a thrilling jDPG com   more ...
FLAMINGOS

Strong Showing at International PLANCKS Competition

The Heidelberg teams performed impressively in Barcelona – securing second place and a solid position in the top half.   more ...
MU3E GROUP ANDRE SCHÖNING

Second funding period for Mu3e experiment

A further period of four years has been granted by the DFG.   more ...
CERN: MAXIMILIEN BRICE

Breakthrough Prize in Fundamental Physics 2025

The 2025 prize has been awarded to ALICE, ATLAS, CMS and LHCb   more ...
FLAMINGOS

First and fifth place in the Dopplers-competition

Congratulations to "Flamingos" and "Knechte Ruprechts"!   more ...

Physics colloquium

Friday, 4. July 2025 5:00 pm  Generative Neural Networks for the Sciences

Prof. Dr. Ullrich Köthe, Interdisciplinary Center for Scientific Computing (IWR), Heidelberg

Generative modelling with normalizing flows has worked well in scientific applications like simulation-based inference. However, the peculiar design makes it difficult to incorporate prior knowledge (such as laws of physics or chemistry) into their architecture. Free-form flows eliminate this restriction by means of a new training algorithm. Manifold free-form flows elegantly exploit these opportunities in the case when we know that the data reside on a manifold. The talk will explain the underlying theory and present experimental evidence for the promising behavior of the new approach.


 

Contact

Dekanat der Fakultät für Physik und Astronomie
Im Neuenheimer Feld 226
69120 Heidelberg

E-Mail: dekanat (at) physik.uni-heidelberg.de

Tel: +49 6221 54 19648