Loss landscapes and optimization in over-parameterized non-linear systems and neural networks

The success of deep learning is due, to a large extent, to the remarkable effectiveness of gradient-based optimization methods applied to large neural networks. The purpose of this work is to propose a modern view and a general mathematical framework for loss landscapes and efficient optimization in...

Celý popis

Podrobná bibliografia

Vydané v:	Applied and computational harmonic analysis Ročník 59; s. 85 - 116
Hlavní autori:	Liu, Chaoyue, Zhu, Libin, Belkin, Mikhail
Médium:	Journal Article
Jazyk:	English
Vydavateľské údaje:	Elsevier Inc 01.07.2022
Predmet:	Deep learning Non-linear optimization Over-parameterized models PL⁎ condition Deep learning PL⁎ condition Over-parameterized models Non-linear optimization
ISSN:	1063-5203
On-line prístup:	Získať plný text
Tagy:	Pridať tag Žiadne tagy, Buďte prvý, kto otaguje tento záznam!