Deep Learning of Part-based Representation of Data Using Sparse Autoencoders with Nonnegativity Constraints

01/12/2016
by   Ehsan Hosseini-Asl, et al.
0

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.

READ FULL TEXT

page 4

page 6

page 7

page 8

page 9

page 11

page 13

research
12/16/2018

Deep Clustering Based on a Mixture of Autoencoders

In this paper we propose a Deep Autoencoder MIxture Clustering (DAMIC) a...
research
01/31/2018

Deep Learning of Constrained Autoencoders for Enhanced Understanding of Data

Unsupervised feature extractors are known to perform an efficient and di...
research
04/13/2019

A joint autoencoder for prediction and its application in GPS trajectory data

In scientific fields, data sparsity greatly affects prediction performan...
research
03/20/2018

Learning the Hierarchical Parts of Objects by Deep Non-Smooth Nonnegative Matrix Factorization

Nonsmooth Nonnegative Matrix Factorization (nsNMF) is capable of produci...
research
02/02/2021

Deep Autoencoder-based Fuzzy C-Means for Topic Detection

Topic detection is a process for determining topics from a collection of...
research
09/08/2020

Dual-constrained Deep Semi-Supervised Coupled Factorization Network with Enriched Prior

Nonnegative matrix factorization is usually powerful for learning the "s...
research
07/09/2018

Using Swarm Optimization To Enhance Autoencoders Images

Autoencoders learn data representations through reconstruction. Robust t...

Please sign up or login with your details

Forgot password? Click here to reset