A Source Domain Extension Method for Inductive Transfer Learning Based on Flipping Output
Transfer learning aims for high accuracy by applying knowledge of source domains for which data collection is easy in order to target domains where data collection is difficult, and has attracted attention in recent years because of its significant potential to enable the application of machine lear...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2019-05-01
|
Series: | Algorithms |
Subjects: | |
Online Access: | https://www.mdpi.com/1999-4893/12/5/95 |
id |
doaj-51597845b2264194b9cb1494e290a7d4 |
---|---|
record_format |
Article |
spelling |
doaj-51597845b2264194b9cb1494e290a7d42020-11-25T01:33:15ZengMDPI AGAlgorithms1999-48932019-05-011259510.3390/a12050095a12050095A Source Domain Extension Method for Inductive Transfer Learning Based on Flipping OutputYasutake Koishi0Shuichi Ishida1Tatsuo Tabaru2Hiroyuki Miyamoto3Advanced Manufacturing Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), Saga 841-0052, JapanAdvanced Manufacturing Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), Saga 841-0052, JapanAdvanced Manufacturing Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), Saga 841-0052, JapanGraduate School of Life Science and Systems Engineering, Kyushu Institute of Technology (Kyutech), Fukuoka 808-0196, JapanTransfer learning aims for high accuracy by applying knowledge of source domains for which data collection is easy in order to target domains where data collection is difficult, and has attracted attention in recent years because of its significant potential to enable the application of machine learning to a wide range of real-world problems. However, since the technique is user-dependent, with data prepared as a source domain which in turn becomes a knowledge source for transfer learning, it often involves the adoption of inappropriate data. In such cases, the accuracy may be reduced due to “negative transfer.” Thus, in this paper, we propose a novel transfer learning method that utilizes the flipping output technique to provide multiple labels in the source domain. The accuracy of the proposed method is statistically demonstrated to be significantly better than that of the conventional transfer learning method, and its effect size is as high as 0.9, showing high performance.https://www.mdpi.com/1999-4893/12/5/95transfer learningensemble learningdata expansionflipping output |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Yasutake Koishi Shuichi Ishida Tatsuo Tabaru Hiroyuki Miyamoto |
spellingShingle |
Yasutake Koishi Shuichi Ishida Tatsuo Tabaru Hiroyuki Miyamoto A Source Domain Extension Method for Inductive Transfer Learning Based on Flipping Output Algorithms transfer learning ensemble learning data expansion flipping output |
author_facet |
Yasutake Koishi Shuichi Ishida Tatsuo Tabaru Hiroyuki Miyamoto |
author_sort |
Yasutake Koishi |
title |
A Source Domain Extension Method for Inductive Transfer Learning Based on Flipping Output |
title_short |
A Source Domain Extension Method for Inductive Transfer Learning Based on Flipping Output |
title_full |
A Source Domain Extension Method for Inductive Transfer Learning Based on Flipping Output |
title_fullStr |
A Source Domain Extension Method for Inductive Transfer Learning Based on Flipping Output |
title_full_unstemmed |
A Source Domain Extension Method for Inductive Transfer Learning Based on Flipping Output |
title_sort |
source domain extension method for inductive transfer learning based on flipping output |
publisher |
MDPI AG |
series |
Algorithms |
issn |
1999-4893 |
publishDate |
2019-05-01 |
description |
Transfer learning aims for high accuracy by applying knowledge of source domains for which data collection is easy in order to target domains where data collection is difficult, and has attracted attention in recent years because of its significant potential to enable the application of machine learning to a wide range of real-world problems. However, since the technique is user-dependent, with data prepared as a source domain which in turn becomes a knowledge source for transfer learning, it often involves the adoption of inappropriate data. In such cases, the accuracy may be reduced due to “negative transfer.” Thus, in this paper, we propose a novel transfer learning method that utilizes the flipping output technique to provide multiple labels in the source domain. The accuracy of the proposed method is statistically demonstrated to be significantly better than that of the conventional transfer learning method, and its effect size is as high as 0.9, showing high performance. |
topic |
transfer learning ensemble learning data expansion flipping output |
url |
https://www.mdpi.com/1999-4893/12/5/95 |
work_keys_str_mv |
AT yasutakekoishi asourcedomainextensionmethodforinductivetransferlearningbasedonflippingoutput AT shuichiishida asourcedomainextensionmethodforinductivetransferlearningbasedonflippingoutput AT tatsuotabaru asourcedomainextensionmethodforinductivetransferlearningbasedonflippingoutput AT hiroyukimiyamoto asourcedomainextensionmethodforinductivetransferlearningbasedonflippingoutput AT yasutakekoishi sourcedomainextensionmethodforinductivetransferlearningbasedonflippingoutput AT shuichiishida sourcedomainextensionmethodforinductivetransferlearningbasedonflippingoutput AT tatsuotabaru sourcedomainextensionmethodforinductivetransferlearningbasedonflippingoutput AT hiroyukimiyamoto sourcedomainextensionmethodforinductivetransferlearningbasedonflippingoutput |
_version_ |
1725078577487020032 |