Deep Learning of Part-based Representation of Data Using Sparse Autoencoders with Nonnegativity Constraints
arXiv:1601.02733 · doi:10.1109/TNNLS.2015.2479223
Abstract
We demonstrate a new deep learning autoencoder network, trained by a nonnegativity constraint algorithm (NCAE), that learns features which 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 dataset. 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.
Accepted for publication in IEEE Transactions of Neural Networks and Learning Systems
References in corpus (2)
Cited by in corpus (11)
- Deep matrix factorizations
- Deep Learning of Nonnegativity-Constrained Autoencoders for Enhanced Understanding of Data
- Alzheimer's Disease Diagnostics by a Deeply Supervised Adaptable 3D Convolutional Network
- Multiplicative update rules for accelerating deep learning training and increasing robustness
- A Variational Autoencoder for Probabilistic Non-Negative Matrix Factorisation
- Basis-MelGAN: Efficient Neural Vocoder Based on Audio Decomposition
- Learning a Deep Part-based Representation by Preserving Data Distribution
- Clustering Bioactive Molecules in 3D Chemical Space with Unsupervised Deep Learning
- Part-based approximations for morphological operators using asymmetric auto-encoders
- Inducing Hierarchical Compositional Model by Sparsifying Generator Network
- Unsupervised Deep Transfer Feature Learning for Medical Image Classification