Adaptive Non-Uniform Timestep Sampling for Diffusion Model Training

As a highly expressive generative model, diffusion models have demonstrated exceptional success across various domains, including image generation, natural language processing, and combinatorial optimization. However, as data distributions grow more complex, training these models to convergence beco...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	arXiv.org 2024-11
Hauptverfasser:	Kim, Myunsoo, Donghyeon Ki, Seong-Woong Shim, Byung-Jun, Lee
Format:	Artikel
Sprache:	eng
Schlagworte:	Adaptive sampling Combinatorial analysis Convergence Diffusion rate Image processing Natural language processing Sampling methods Speech recognition
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!