We propose an optimization method obtained by the approximation of a novel discretization approach for gradient dynamics recently proposed by the authors. It is shown that the proposed algorithm ensures convergence for all amplitudes of the step size, contrarily to classical implementations.
A gradient descent algorithm built on approximate discrete gradients / Moreschini, A.; Mattioni, M.; Monaco, S.; Normand-Cyrot, D.. - (2022), pp. 343-348. (Intervento presentato al convegno 26th International Conference on System Theory, Control and Computing, ICSTCC 2022 tenutosi a Sinaia; Romania) [10.1109/ICSTCC55426.2022.9931872].
A gradient descent algorithm built on approximate discrete gradients
Moreschini A.
;Mattioni M.;Monaco S.;
2022
Abstract
We propose an optimization method obtained by the approximation of a novel discretization approach for gradient dynamics recently proposed by the authors. It is shown that the proposed algorithm ensures convergence for all amplitudes of the step size, contrarily to classical implementations.| File | Dimensione | Formato | |
|---|---|---|---|
|
Moreschini_AGradientDescent_2022.pdf
solo gestori archivio
Tipologia:
Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza:
Tutti i diritti riservati (All rights reserved)
Dimensione
779.93 kB
Formato
Adobe PDF
|
779.93 kB | Adobe PDF | Contatta l'autore |
|
Moreschini_AGradientDescent_2022_Preprint.pdf
accesso aperto
Tipologia:
Documento in Pre-print (manoscritto inviato all'editore, precedente alla peer review)
Licenza:
Creative commons
Dimensione
655.1 kB
Formato
Adobe PDF
|
655.1 kB | Adobe PDF |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
