This article describes how a NAND memory device is adapted to the implementation of an artificial neural network. To this aim, a new hardware realization of the elementary synapse (called here "Synaptic Cell") is proposed. Furthermore, a method for connecting neurons to form a multilayer ANN is proposed with necessary algorithms to run it. Some examples of neural behavior, like function approximation and classification are reported.
A neural network implemented on NAND memory / Minucci, U.; Santis, L. D.; Irrera, F.; Vali, T.. - (2020), pp. 1-4. (Intervento presentato al convegno 2020 IEEE International Memory Workshop, IMW 2020 tenutosi a deu) [10.1109/IMW48823.2020.9108138].
A neural network implemented on NAND memory
Irrera F.;
2020
Abstract
This article describes how a NAND memory device is adapted to the implementation of an artificial neural network. To this aim, a new hardware realization of the elementary synapse (called here "Synaptic Cell") is proposed. Furthermore, a method for connecting neurons to form a multilayer ANN is proposed with necessary algorithms to run it. Some examples of neural behavior, like function approximation and classification are reported.| File | Dimensione | Formato | |
|---|---|---|---|
|
Minucci_Neural-network_2020.pdf
solo gestori archivio
Tipologia:
Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza:
Tutti i diritti riservati (All rights reserved)
Dimensione
2.98 MB
Formato
Adobe PDF
|
2.98 MB | Adobe PDF | Contatta l'autore |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
