On the existence of global minima and convergence analyses for gradient descent methods in the training of deep neural networks
Jentzen A, Riekert A
Research article (journal) | Peer reviewedDetails about the publication
Journal: Journal of Machine Learning (J. Mach. Learn.)
Volume: 1
Issue: 2
Page range: 141-246
Status: Published
Release year: 2022
Language in which the publication is written: English
DOI: 10.4208/jml.220114a
Link to the full text: https://global-sci.org/intro/article_detail/jml/20801.html
Keywords: Deep learning; Global minimum; Kurdyka-Łojasiewicz inequality; Gradient descent; Optimization
Authors from the University of Münster
| Jentzen, Arnulf | Institute for Analysis and Numerics |
| Riekert, Adrian | Mathematical Institute |