Locality-constrained transfer coding for heterogeneous domain adaptation

Li, Jingjing, Lu, Ke, Zhu, Lei and Li, Zhihui (2017). Locality-constrained transfer coding for heterogeneous domain adaptation. In: 28th Australasian Database Conference, ADC 2017, Brisbane, QLD, Australia, (193-204). 25-28 September 2017. doi:10.1007/978-3-319-68155-9_15

Author Li, Jingjing
Lu, Ke
Zhu, Lei
Li, Zhihui
Title of paper Locality-constrained transfer coding for heterogeneous domain adaptation
Conference name 28th Australasian Database Conference, ADC 2017
Conference location Brisbane, QLD, Australia
Conference dates 25-28 September 2017
Journal name Lecture Notes in Computer Science   Check publisher's open access policy
Place of Publication Cham, Switzerland
Publisher Springer Nature
Publication Year 2017
Sub-type Fully published paper
DOI 10.1007/978-3-319-68155-9_15
Open Access Status Not yet assessed
ISBN 9783319681542
ISSN 1611-3349
Volume 10538
Start page 193
End page 204
Total pages 12
Chapter number 15
Total chapters 22
Language eng
Abstract/Summary Currently, most of widely used databases are label-wise. In other words, people organize their data with corresponding labels, e.g., class information, keywords and description, for the convenience of indexing and retrieving. However, labels of the data from a novel application usually are not available, and labeling by hand is very expensive. To address this, we propose a novel approach based on transfer learning. Specifically, we aim at tackling heterogeneous domain adaptation (HDA). HDA is a crucial topic in transfer learning. Two inevitable issues, feature discrepancy and distribution divergence, get in the way of HDA. However, due to the significant challenges of HDA, previous work commonly focus on handling one of them and neglect the other. Here we propose to deploy locality-constrained transfer coding (LCTC) to simultaneously alleviate the feature discrepancy and mitigate the distribution divergence. Our method is powered by two tactics: feature alignment and distribution alignment. The former learns new transferable feature representations by sharing-dictionary coding and the latter aligns the distribution gaps on the new feature space. By formulating the problem into a unified objective and optimizing it via an iterative fashion, the two tactics are reinforced by each other and the two domains are drawn closer under the new representations. Extensive experiments on image classification and text categorization verify the superiority of our method against several state-of-the-art approaches.
Keyword Domain adaptation
Knowledge discovery
Transfer learning
Q-Index Code E1
Q-Index Status Provisional Code
Institutional Status UQ

Document type: Conference Paper
Sub-type: Fully published paper
Collections: HERDC Pre-Audit
School of Information Technology and Electrical Engineering Publications
Version Filter Type
Citation counts: Scopus Citation Count Cited 0 times in Scopus Article
Google Scholar Search Google Scholar
Created: Sun, 17 Dec 2017, 00:38:02 EST