Cargando…

Compression of a Deep Competitive Network Based on Mutual Information for Underwater Acoustic Targets Recognition

The accuracy of underwater acoustic targets recognition via limited ship radiated noise can be improved by a deep neural network trained with a large number of unlabeled samples. However, redundant features learned by deep neural network have negative effects on recognition accuracy and efficiency....

Descripción completa

Detalles Bibliográficos
Autores principales: Shen, Sheng, Yang, Honghui, Sheng, Meiping
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2018
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7512758/
https://www.ncbi.nlm.nih.gov/pubmed/33265334
http://dx.doi.org/10.3390/e20040243
_version_ 1783586231905419264
author Shen, Sheng
Yang, Honghui
Sheng, Meiping
author_facet Shen, Sheng
Yang, Honghui
Sheng, Meiping
author_sort Shen, Sheng
collection PubMed
description The accuracy of underwater acoustic targets recognition via limited ship radiated noise can be improved by a deep neural network trained with a large number of unlabeled samples. However, redundant features learned by deep neural network have negative effects on recognition accuracy and efficiency. A compressed deep competitive network is proposed to learn and extract features from ship radiated noise. The core idea of the algorithm includes: (1) Competitive learning: By integrating competitive learning into the restricted Boltzmann machine learning algorithm, the hidden units could share the weights in each predefined group; (2) Network pruning: The pruning based on mutual information is deployed to remove the redundant parameters and further compress the network. Experiments based on real ship radiated noise show that the network can increase recognition accuracy with fewer informative features. The compressed deep competitive network can achieve a classification accuracy of [Formula: see text] , which is [Formula: see text] higher than deep competitive network and [Formula: see text] higher than the state-of-the-art signal processing feature extraction methods.
format Online
Article
Text
id pubmed-7512758
institution National Center for Biotechnology Information
language English
publishDate 2018
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-75127582020-11-09 Compression of a Deep Competitive Network Based on Mutual Information for Underwater Acoustic Targets Recognition Shen, Sheng Yang, Honghui Sheng, Meiping Entropy (Basel) Article The accuracy of underwater acoustic targets recognition via limited ship radiated noise can be improved by a deep neural network trained with a large number of unlabeled samples. However, redundant features learned by deep neural network have negative effects on recognition accuracy and efficiency. A compressed deep competitive network is proposed to learn and extract features from ship radiated noise. The core idea of the algorithm includes: (1) Competitive learning: By integrating competitive learning into the restricted Boltzmann machine learning algorithm, the hidden units could share the weights in each predefined group; (2) Network pruning: The pruning based on mutual information is deployed to remove the redundant parameters and further compress the network. Experiments based on real ship radiated noise show that the network can increase recognition accuracy with fewer informative features. The compressed deep competitive network can achieve a classification accuracy of [Formula: see text] , which is [Formula: see text] higher than deep competitive network and [Formula: see text] higher than the state-of-the-art signal processing feature extraction methods. MDPI 2018-04-02 /pmc/articles/PMC7512758/ /pubmed/33265334 http://dx.doi.org/10.3390/e20040243 Text en © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Shen, Sheng
Yang, Honghui
Sheng, Meiping
Compression of a Deep Competitive Network Based on Mutual Information for Underwater Acoustic Targets Recognition
title Compression of a Deep Competitive Network Based on Mutual Information for Underwater Acoustic Targets Recognition
title_full Compression of a Deep Competitive Network Based on Mutual Information for Underwater Acoustic Targets Recognition
title_fullStr Compression of a Deep Competitive Network Based on Mutual Information for Underwater Acoustic Targets Recognition
title_full_unstemmed Compression of a Deep Competitive Network Based on Mutual Information for Underwater Acoustic Targets Recognition
title_short Compression of a Deep Competitive Network Based on Mutual Information for Underwater Acoustic Targets Recognition
title_sort compression of a deep competitive network based on mutual information for underwater acoustic targets recognition
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7512758/
https://www.ncbi.nlm.nih.gov/pubmed/33265334
http://dx.doi.org/10.3390/e20040243
work_keys_str_mv AT shensheng compressionofadeepcompetitivenetworkbasedonmutualinformationforunderwateracoustictargetsrecognition
AT yanghonghui compressionofadeepcompetitivenetworkbasedonmutualinformationforunderwateracoustictargetsrecognition
AT shengmeiping compressionofadeepcompetitivenetworkbasedonmutualinformationforunderwateracoustictargetsrecognition