Document Detail


Domain Transfer Multiple Kernel Learning.
MedLine Citation:
PMID:  21646679     Owner:  NLM     Status:  Publisher    
Abstract/OtherAbstract:
Cross-domain learning methods have shown promising results by leveraging labeled patterns from the auxiliary domain to learn a robust classifier for the target domain which has only a limited number of labeled samples. To cope with the considerable change between feature distributions of different domains, we propose a new cross-domain kernel learning framework into which many existing kernel methods can be readily incorporated. Our framework, referred to as Domain Transfer Multiple Kernel Learning (DTMKL), simultaneously learns a kernel function and a robust classifier by minimizing both the structural risk functional and the distribution mismatch between labeled and unlabeled samples from the auxiliary and target domains. Under the DTMKL framework, we also propose two novel methods by using SVM and pre-learned classifiers, respectively. Comprehensive experiments on three domain adaptation data sets (i.e., TRECVID, 20 Newsgroups and email spam data sets) demonstrate that DTMKL based methods outperform existing cross-domain learning and multiple kernel learning methods.
Authors:
Lixin Duan; Ivor W Tsang; Dong Xu
Related Documents :
20735249 - Astrobiological benefits of human space exploration.
7599589 - Building and maintaining a library gopher: traditional skills applied to emerging resou...
8666469 - The effects of a life review program on disorientation, social interaction and self-est...
8273959 - Intraosseous infusion: success of a standardized regional training program for prehospi...
22132279 - Automated data mining: an innovative and efficient web-based approach to maintaining re...
22081549 - Self-directed student research through analysis of microarray datasets: a computer-base...
1918759 - Nutrition information needs during cardiac rehabilitation: perceptions of the cardiac p...
6691699 - Part-time higher professional training: an analysis of content.
16122619 - Clinicians' information sources for new substance abuse treatment.
Publication Detail:
Type:  JOURNAL ARTICLE     Date:  2011-5-26
Journal Detail:
Title:  IEEE transactions on pattern analysis and machine intelligence     Volume:  -     ISSN:  1939-3539     ISO Abbreviation:  -     Publication Date:  2011 May 
Date Detail:
Created Date:  2011-6-7     Completed Date:  -     Revised Date:  -    
Medline Journal Info:
Nlm Unique ID:  9885960     Medline TA:  IEEE Trans Pattern Anal Mach Intell     Country:  -    
Other Details:
Languages:  ENG     Pagination:  -     Citation Subset:  -    
Affiliation:
Nanyang Technological University, Singapore.
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine


Previous Document:  Scalable Face Image Retrieval with Identity-Based Quantization and Multi-Reference Re-ranking.
Next Document:  Towards a Practical Face Recognition System: Robust Alignment and Illumination by Sparse Representat...