Reducing Memory and Computational Cost for Deep Neural Network Training with Quantized Parameter Updates.

Saved in:

Bibliographic Details
Title:	Reducing Memory and Computational Cost for Deep Neural Network Training with Quantized Parameter Updates.
Authors:	Buron, Leo¹ leo.buron@uni-due.de, Erbslöh, Andreas¹ andreas.erbsloeh@uni-due.de, Schiele, Gregor¹ gregor.schiele@uni-due.de
Source:	Journal of Universal Computer Science (JUCS). 2025, Vol. 31 Issue 9, p963-979. 17p.
Database:	Mathematics Source

Description
ISSN:	0948695X
DOI:	10.3897/jucs.164737