Поиск

Автор

Ключевое слово

Тип статьи

DOI

ORCID

Дополнительно:

Период публикации

Репозиторий

По вашему запросу найдено документов: 57451

Страница 10 из 5746

In this paper, an issue of stability of fully connected neural network (perceptron), forecasting

is proposed, within which a method of dynamic local approximation using neural network models (DLANN) has been

of speech. After comparing traditional classification methods and neural network classification methods

of building a predicting neural-net model of the time sequence of network traffic. Topological data analysis

of social media data. Neural network modeling, statistical analysis, and differential equations were used

and show improvements for modem neural network NLP models on standard tasks such as sentiment analysis.

of seismic data can be significantly enhanced through the use of neural network modeling, attribute analysis

synthesis problem. Feed-forward neural network was used as an approximation of the control function

. In this paper, the problem of building a neural network identification model for oil pipeline transportation has

neither deep neural networks, which have been successfully applied to ad-hoc information retrieval and Web

Страница 10 из 5746