Deep Learning of Part-Based Representation of Data Using Sparse Autoencoders With Nonnegativity Constraints | IEEE Journals & Magazine | IEEE Xplore

Deep Learning of Part-Based Representation of Data Using Sparse Autoencoders With Nonnegativity Constraints


Abstract:

We demonstrate a new deep learning autoencoder network, trained by a nonnegativity constraint algorithm (nonnegativity-constrained autoencoder), that learns features that...Show More

Abstract:

We demonstrate a new deep learning autoencoder network, trained by a nonnegativity constraint algorithm (nonnegativity-constrained autoencoder), that learns features that show part-based representation of data. The learning algorithm is based on constraining negative weights. The performance of the algorithm is assessed based on decomposing data into parts and its prediction performance is tested on three standard image data sets and one text data set. The results indicate that the nonnegativity constraint forces the autoencoder to learn features that amount to a part-based representation of data, while improving sparsity and reconstruction quality in comparison with the traditional sparse autoencoder and nonnegative matrix factorization. It is also shown that this newly acquired representation improves the prediction performance of a deep neural network.
Published in: IEEE Transactions on Neural Networks and Learning Systems ( Volume: 27, Issue: 12, December 2016)
Page(s): 2486 - 2498
Date of Publication: 28 October 2015

ISSN Information:

PubMed ID: 26529786

Funding Agency:


Contact IEEE to Subscribe

References

References is not available for this document.