52647

Автор(ы): 

Автор(ов): 

4

Параметры публикации

Тип публикации: 

Статья в журнале/сборнике

Название: 

Projectional Learning Laws for Differential Neural Networks Based on Double-Averaged Sub-Gradient Descent Technique

ISBN/ISSN: 

0302-9743

DOI: 

10.1007/978-3-030-22796-8

Наименование источника: 

  • Lecture Notes in Computer Science

Обозначение и номер тома: 

vol 11554

Город: 

  • Cham, Switzerland

Издательство: 

  • Springer Link

Год издания: 

2019

Страницы: 

28-38
Аннотация
Abstract. A new method to design learning laws for neural networks with continuous dynamics is proposed in this study. The learning method is based on the so-called double-averaged descendent technique (DASGDT), which is a variant of the gradient-descendent method. The learning law implements a double averaged algorithm which filters the effect of uncertainties of the states, which are continuously measurable. The learning law overcomes the classical assumption on the strict convexity of the functional with respect to the weights. The photocatalytic ozonation process of a single contaminant is estimated using the learning law design proposed in this study.

Библиографическая ссылка: 

Чаирес И.И., Позняк А.С., Назин А.В., Позняк Т.И. Projectional Learning Laws for Differential Neural Networks Based on Double-Averaged Sub-Gradient Descent Technique // Lecture Notes in Computer Science. 2019. vol 11554. С. 28-38.