Next Article in Journal / Special Issue
Using the Guided Fireworks Algorithm for Local Backlight Dimming
Previous Article in Journal
Stability and Heat Input Controllability of Two Different Modulations for Double-Pulse MIG Welding
Previous Article in Special Issue
Accelerating Image Classification using Feature Map Similarity in Convolutional Neural Networks
Article Menu
Issue 1 (January-1) cover image

Export Article

Open AccessArticle

Improvement in Classification Performance Based on Target Vector Modification for All-Transfer Deep Learning

1
Technology Innovation Division, Panasonic Corporation, Tokyo 135-8072, Japan
2
Business Innovation Division, Panasonic Corporation, Osaka 570-8501, Japan
3
Department of Life Science and Technology, Tokyo Institute of Technology, Tokyo 152-8550, Japan
*
Author to whom correspondence should be addressed.
Appl. Sci. 2019, 9(1), 128; https://doi.org/10.3390/app9010128
Received: 22 October 2018 / Revised: 29 November 2018 / Accepted: 26 December 2018 / Published: 1 January 2019
(This article belongs to the Special Issue Advanced Intelligent Imaging Technology)
  |  
PDF [7180 KB, uploaded 1 January 2019]
  |  

Abstract

This paper proposes a target vector modification method for the all-transfer deep learning (ATDL) method. Deep neural networks (DNNs) have been used widely in many applications; however, the DNN has been known to be problematic when large amounts of training data are not available. Transfer learning can provide a solution to this problem. Previous methods regularize all layers, including the output layer, by estimating the relation vectors, which are then used instead of one-hot target vectors of the target domain. These vectors are estimated by averaging the target domain data of each target domain label in the output space. This method improves the classification performance, but it does not consider the relation between the relation vectors. From this point of view, we propose a relation vector modification based on constrained pairwise repulsive forces. High pairwise repulsive forces provide large distances between the relation vectors. In addition, the risk of divergence is mitigated by the constraint based on distributions of the output vectors of the target domain data. We apply our method to two simulation experiments and a disease classification using two-dimensional electrophoresis images. The experimental results show that reusing all layers through our estimation method is effective, especially for a significantly small number of the target domain data. View Full-Text
Keywords: deep neural network; transfer learning; proteomics; sepsis classification deep neural network; transfer learning; proteomics; sepsis classification
Figures

Graphical abstract

This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).
SciFeed

Share & Cite This Article

MDPI and ACS Style

Sawada, Y.; Sato, Y.; Nakada, T.; Yamaguchi, S.; Ujimoto, K.; Hayashi, N. Improvement in Classification Performance Based on Target Vector Modification for All-Transfer Deep Learning. Appl. Sci. 2019, 9, 128.

Show more citation formats Show less citations formats

Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Related Articles

Article Metrics

Article Access Statistics

1

Comments

[Return to top]
Appl. Sci. EISSN 2076-3417 Published by MDPI AG, Basel, Switzerland RSS E-Mail Table of Contents Alert
Back to Top