The gradient analysis for neural networks with veyletdecomposition of a target vector

Authors

  • В. И. Дрововозов Национальный авиационный университет
  • А. К. Хемраев Национальный авиационный университет

DOI:

https://doi.org/10.18372/2073-4751.3.7075

Abstract

In this article presents an analysis of the gradient for some cases of neural networks with wavelet and wavelet-like decomposition of the target vector - a new type of neural network specialized in speech recognition and signal conversion, and to expedite the speed and quality of education compared to the standard perceptron. Through this analysis it is shown that in a sufficiently broad framework of neural networks with wavelet decomposition of the target vector better standard multilayer perceptron

Author Biography

В. И. Дрововозов, Национальный авиационный университет

к.т.н.

References

Tebelskis, J. Speech Recognition using Neural Networks: PhD thesis, Doctor of Philosophy in Computer Science/ Joe Tebelskis; School of Computer Science, Carnegie Mellon University.– Pittsburgh, Pennsylvania. – 1995.– 179 р.

Handbook of neural network signal processing/ Edited by Yu Hen Hu, Jenq-Neng Hwang. – Boca Raton; London; New York, Washington D.C.: CRC press. – 2001. – 384 р.

Ф.Г. Бойков Применение вейвлетанализа в задачах автоматического распознавания речи: Дис. кандидата физикоматематических наук: 05.13.18 / Фѐдор Геннадьевич Бойков. – М., 2003. – 111 с.

Добеши И. Десять лекций по вейвлетам.– Ижевск: НИЦ «Регулярная и хаотическая динамика», – 2001. – 464 с.

Астапов К.А. Применение вейвлетпреобразования для сокращения области значения искусственных нейронных сетей на примере задачи распознавания речи // Электронный научно-инновационный журнал «Инженерный вестник Дона», Ростов-на-Дону: – 2009. – № 1. – С. 24-29.

Червяков Н.И., Астапов К.А. Использование вейвлетов для улучшения параметров нейронных сетей в задачах распознавания речи. // Инфокоммуникационные технологии – №°4 (2008) – Самара: Издательство ПГУТИ, – 2008. – С. 9-12.

Downloads

Published

2012-09-26

How to Cite

Дрововозов, В. И., & Хемраев, А. К. (2012). The gradient analysis for neural networks with veyletdecomposition of a target vector. Problems of Informatization and Control, 3(39), 43–46. https://doi.org/10.18372/2073-4751.3.7075

Issue

Vol. 3 No. 39 (2012)

Section

Статті

License

The scientific journal adheres to the principles of Open Access and provides free, immediate, and permanent access to all published materials without financial, technical, or legal barriers for readers.

All articles are published in Open Access under the Creative Commons Attribution 4.0 International (CC BY 4.0) license.

Copyright

Authors who publish their works in the journal:

  • retain the copyright to their publications;

  • grant the journal the right of first publication of the article;

  • agree to the distribution of their materials under the CC BY 4.0 license;

  • have the right to reuse, archive, and distribute their works (including in institutional and subject repositories), provided that proper reference is made to the original publication in the journal.