Evaluation of storage and functioning characteristics of artificial neural networks on the basis of a neurocomputer device

Autor:	V A Romanchuk
Rok vydání:	2020
Předmět:	Artificial neural network Basis (linear algebra) Computer science business.industry Artificial intelligence business
Zdroj:	IOP Conference Series: Materials Science and Engineering. 734:012009
ISSN:	1757-899X 1757-8981
DOI:	10.1088/1757-899x/734/1/012009
Popis:	A solution to the problem of evaluating the temporal and quantitative characteristics of the storage and processing of artificial neural networks based on a neurocomputer device is proposed. The most popular and used topologies of artificial neural networks are considered, for which analytical relationships are presented that allow evaluating the training cycle of an artificial neural network, the amount of memory needed and the amount of transmitted data. The proposed relationships differ in that approaches and characteristics unique to neuroprocessor devices are offered for evaluation when implementing the calculations presented in the neural network logical basis.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::f9e2a92be9c74148ba62daac7d313add https://doi.org/10.1088/1757-899x/734/1/012009 Zobrazit plný text záznamu