Research on denoising sparse autoencoder

Autoencoder can learn the structure of data adaptively and represent data efficiently. These properties make autoencoder not only suit huge volume and variety of data well but also overcome expensive designing cost and poor generalization. Moreover, using autoencoder in deep learning to implement fe...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:International journal of machine learning and cybernetics Ročník 8; číslo 5; s. 1719 - 1729
Hlavní autori: Meng, Lingheng, Ding, Shifei, Xue, Yu
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Berlin/Heidelberg Springer Berlin Heidelberg 01.10.2017
Springer Nature B.V
Predmet:
Artificial Intelligence
Classification
Complex Systems
Computational Intelligence
Control
Engineering
Entropy
Feature extraction
Machine learning
Mechatronics
Neural networks
Noise reduction
Original Article
Pattern Recognition
Robotics
Robustness
Sparsity
Systems Biology
Feature extraction
Sparse coding
Autoencoder
Unsupervised learning
Deep networks
ISSN:1868-8071, 1868-808X
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:Autoencoder can learn the structure of data adaptively and represent data efficiently. These properties make autoencoder not only suit huge volume and variety of data well but also overcome expensive designing cost and poor generalization. Moreover, using autoencoder in deep learning to implement feature extraction could draw better classification accuracy. However, there exist poor robustness and overfitting problems when utilizing autoencoder. In order to extract useful features, meanwhile improve robustness and overcome overfitting, we studied denoising sparse autoencoder through adding corrupting operation and sparsity constraint to traditional autoencoder. The results suggest that different autoencoders mentioned in this paper have some close relation and the model we researched can extract interesting features which can reconstruct original data well. In addition, all results show a promising approach to utilizing the proposed autoencoder to build deep models.
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1868-8071
1868-808X
DOI:10.1007/s13042-016-0550-y