Parts | Days | Selection | Search | Updates | Downloads | Help

AKPIK: Arbeitskreis Physik, moderne Informationstechnologie und Künstliche Intelligenz

AKPIK 4: AKPIK III: Simulation & Application

AKPIK 4.5: Talk

Thursday, March 18, 2021, 17:00–17:15, AKPIKa

How normalizing flows generalize the Gaussian distribution — •Thorsten Glüsenkamp — FAU Nürnberg, Erlangen, Germany

Normalizing-flows, flexible continuous probability distributions, are emerging in many machine-learning applications. We discuss how they can be described as generalizations of Multivariate Normal distributions and how this might guide us for the choice of which flow to use.