Allgemeine Informationen

Veranstaltungsname	Seminar: Blockseminar: Neural ODE and Generative Modelling für Master
Veranstaltungsnummer	MTH-1360; MTH-1410
Semester	SS 2025
Aktuelle Anzahl der Teilnehmenden	0
Heimat-Einrichtung	Mathematische Bildverarbeitung
Veranstaltungstyp	Seminar in der Kategorie Lehre
Veranstaltung findet in Präsenz statt / hat Präsenz-Bestandteile	Ja
Hauptunterrichtssprache	englisch

Lehrende

Prof.Dr. Stefania Petra

Räume und Zeiten

Keine Raumangabe

Studienbereiche

Kommentar/Beschreibung

Score-based generative models have demonstrated state-of-the-art performance in numerous applications in recent years. The central concept behind these models involves the gradual injection of noise into the training data, followed by learning the reverse process to generate new samples. The training and sampling procedures can be conducted independently. The learning phase is facilitated by noise-conditional score networks, while sampling can be accomplished through various methods, including Langevin Monte Carlo approaches, stochastic differential equations, ordinary differential equations, and various combinations.

This seminar will overview generative models and common architectures, comparing different training objectives and sampling methods. We will examine normalizing flows, score matching, and Langevin dynamics. Additionally, we will discuss the use of stochastic differential equations (SDEs) in score-based models and conclude with comparisons to other diffusion models and potential enhancements in sample generation.

Seminar: Blockseminar: Neural ODE and Generative Modelling für Master - Details