Multi-Valued Autoencoders and Classification of Large-Scale Multi-Class Problem

Abstract

Two-layered neural networks are well known as autoencoders (AEs) in order to reduce the dimensionality of data. AEs are successfully employed as pre-trained layers of neural networks for classification tasks. Most of the existing studies conceived real-valued AEs in real-valued neural networks. This study investigated complex- and quaternion-valued AEs for complex- and quaternion-valued neural networks. Inputs, weights, biases, and outputs in complex-valued AE (CAE) are complex variables, whereas those in quaternion-valued AE (QAE) are quaternions. In both methods, a split-type activation function is used in the hidden and output units. To deal with the images using the proposed methods, pairs of pixels are allotted to complex-valued inputs in the CAE and quartets of pixels are allotted to quaternion-valued inputs in the QAE. Proposed autoencoders are tested and performance compared with conventional AE for several tasks which are encoding/decoding, handwritten numeral recognition and large-scale multi-class classification. Proposed CAE and QAE revealed as good recognition methods for the tasks and outperformed conventional AE with significance performance in case of large-scale multi-class images recognition.

Authors and Affiliations

Ryusuke Hata, M. A. H. Akhand, Kazuyuki Murase

Keywords

Related Articles

Secure Medical Internet of Things Framework based on Parkerian Hexad Model

Medical Internet of Things (MIoT) applications enhance medical services by collecting data using devices connected to the IoT. The collected data, which may include personal data and location, is transmitted to mobile de...

Content Analysis of Privacy Management Features in Geosocial Networking Application

Geosocial networking application allows user to share information and communicate with other people within a virtual neighborhood or community. Although most geosocial networking application include privacy management fe...

OntoDI: The Methodology for Ontology Development on Data Integration

The implementations of data integration in current days have many issues to be solved. Heterogeneity of data with non-standardization data, data conflicts between various data sources, data with a different representatio...

Evaluation of APi Interface Design by Applying Cognitive Walkthrough

The usability evaluation of APi interface design was conducted through Cognitive Walkthrough method. APi is a mobile application game designed specifically for preschool children of Tabika Kemas Kampung Berawan, Limbang...

Detecting Public Sentiment of Medicine by Mining Twitter Data

The paper presents a computational method that mines, processes and analyzes Twitter data for detecting public sentiment of medicine. Self-reported patient data are collected over a period of three months by mining the T...

Download PDF file
  • EP ID EP240416
  • DOI 10.14569/IJACSA.2017.081103
  • Views 101
  • Downloads 0

How To Cite

Ryusuke Hata, M. A. H. Akhand, Kazuyuki Murase (2017). Multi-Valued Autoencoders and Classification of Large-Scale Multi-Class Problem. International Journal of Advanced Computer Science & Applications, 8(11), 19-26. https://europub.co.uk/articles/-A-240416