A general domain specific feature transfer framework for hybrid domain adaptation
Heterogeneous domain adaptation needs supplementary information to link up different domains. However, such supplementary information may not always be available in real cases. In this paper, a new problem setting called hybrid domain adaptation is investigated. It is a special case of heterogeneous...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2020
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/138954 |
| _version_ | 1826125714103992320 |
|---|---|
| author | Wei, Pengfei Ke, Yiping Goh, Chi Keong |
| author2 | School of Computer Science and Engineering |
| author_facet | School of Computer Science and Engineering Wei, Pengfei Ke, Yiping Goh, Chi Keong |
| author_sort | Wei, Pengfei |
| collection | NTU |
| description | Heterogeneous domain adaptation needs supplementary information to link up different domains. However, such supplementary information may not always be available in real cases. In this paper, a new problem setting called hybrid domain adaptation is investigated. It is a special case of heterogeneous domain adaptation, in which different domains share some common features, but also have their own domain specific features. We leverage upon common features instead of supplementary information to achieve effective adaptation. We propose a general domain specific feature transfer framework, which can link up different domains using common features and simultaneously reduce domain divergences. Specifically, we learn the translations between common features and domain specific features. Then, we cross-use the learned translations to transfer the domain specific features of one domain to another domain. Finally, we compose a homogeneous space in which the domain divergences are minimized. We instantiate the general framework to a linear case and a nonlinear case. Extensive experiments verify the effectiveness of the two cases. |
| first_indexed | 2024-10-01T06:40:51Z |
| format | Journal Article |
| id | ntu-10356/138954 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T06:40:51Z |
| publishDate | 2020 |
| record_format | dspace |
| spelling | ntu-10356/1389542020-05-14T05:26:09Z A general domain specific feature transfer framework for hybrid domain adaptation Wei, Pengfei Ke, Yiping Goh, Chi Keong School of Computer Science and Engineering Engineering::Computer science and engineering Knowledge Transfer Domain Specific Feature Heterogeneous domain adaptation needs supplementary information to link up different domains. However, such supplementary information may not always be available in real cases. In this paper, a new problem setting called hybrid domain adaptation is investigated. It is a special case of heterogeneous domain adaptation, in which different domains share some common features, but also have their own domain specific features. We leverage upon common features instead of supplementary information to achieve effective adaptation. We propose a general domain specific feature transfer framework, which can link up different domains using common features and simultaneously reduce domain divergences. Specifically, we learn the translations between common features and domain specific features. Then, we cross-use the learned translations to transfer the domain specific features of one domain to another domain. Finally, we compose a homogeneous space in which the domain divergences are minimized. We instantiate the general framework to a linear case and a nonlinear case. Extensive experiments verify the effectiveness of the two cases. NRF (Natl Research Foundation, S’pore) MOE (Min. of Education, S’pore) Accepted version 2020-05-14T05:26:09Z 2020-05-14T05:26:09Z 2019 Journal Article Wei, P., Ke, Y., & Goh, C. K. (2018). A general domain specific feature transfer framework for hybrid domain adaptation. IEEE Transactions on Knowledge and Data Engineering, 31(8), 1440-1451. doi:10.1109/TKDE.2018.2864732 1041-4347 https://hdl.handle.net/10356/138954 10.1109/TKDE.2018.2864732 2-s2.0-85052575719 8 31 1440 1451 en IEEE Transactions on Knowledge and Data Engineering © 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. The published version is available at: https://doi.org/10.1109/TKDE.2018.2864732. application/pdf |
| spellingShingle | Engineering::Computer science and engineering Knowledge Transfer Domain Specific Feature Wei, Pengfei Ke, Yiping Goh, Chi Keong A general domain specific feature transfer framework for hybrid domain adaptation |
| title | A general domain specific feature transfer framework for hybrid domain adaptation |
| title_full | A general domain specific feature transfer framework for hybrid domain adaptation |
| title_fullStr | A general domain specific feature transfer framework for hybrid domain adaptation |
| title_full_unstemmed | A general domain specific feature transfer framework for hybrid domain adaptation |
| title_short | A general domain specific feature transfer framework for hybrid domain adaptation |
| title_sort | general domain specific feature transfer framework for hybrid domain adaptation |
| topic | Engineering::Computer science and engineering Knowledge Transfer Domain Specific Feature |
| url | https://hdl.handle.net/10356/138954 |
| work_keys_str_mv | AT weipengfei ageneraldomainspecificfeaturetransferframeworkforhybriddomainadaptation AT keyiping ageneraldomainspecificfeaturetransferframeworkforhybriddomainadaptation AT gohchikeong ageneraldomainspecificfeaturetransferframeworkforhybriddomainadaptation AT weipengfei generaldomainspecificfeaturetransferframeworkforhybriddomainadaptation AT keyiping generaldomainspecificfeaturetransferframeworkforhybriddomainadaptation AT gohchikeong generaldomainspecificfeaturetransferframeworkforhybriddomainadaptation |