Multi-class heterogeneous domain adaptation
A crucial issue in heterogeneous domain adaptation (HDA) is the ability to learn a feature mapping between different types of features across domains. Inspired by language translation, a word translated from one language corresponds to only a few words in another language, we present an efficient me...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/85717 http://hdl.handle.net/10220/49259 http://jmlr.org/papers/v20/13-580.html |
| _version_ | 1826114579979042816 |
|---|---|
| author | Zhou, Joey Tianyi Tsang, Ivor W. Pan, Sinno Jialin Tan, Mingkui |
| author2 | School of Computer Science and Engineering |
| author_facet | School of Computer Science and Engineering Zhou, Joey Tianyi Tsang, Ivor W. Pan, Sinno Jialin Tan, Mingkui |
| author_sort | Zhou, Joey Tianyi |
| collection | NTU |
| description | A crucial issue in heterogeneous domain adaptation (HDA) is the ability to learn a feature mapping between different types of features across domains. Inspired by language translation, a word translated from one language corresponds to only a few words in another language, we present an efficient method named Sparse Heterogeneous Feature Representation (SHFR) in this paper for multi-class HDA to learn a sparse feature transformation between domains with multiple classes. Specifically, we formulate the problem of learning the feature transformation as a compressed sensing problem by building multiple binary classifiers in the target domain as various measurement sensors, which are decomposed from the target multi-class classification problem. We show that the estimation error of the learned transformation decreases with the increasing number of binary classifiers. In other words, for adaptation across heterogeneous domains to be successful, it is necessary to construct a sufficient number of incoherent binary classifiers from the original multi-class classification problem. To achieve this, we propose to apply the error correcting output correcting (ECOC) scheme to generate incoherent classifiers. To speed up the learning of the feature transformation across domains, we apply an efficient batch-mode algorithm to solve the resultant nonnegative sparse recovery problem. Theoretically, we present a generalization error bound of our proposed HDA method under a multi-class setting. Lastly, we conduct extensive experiments on both synthetic and real-world datasets to demonstrate the superiority of our proposed method over existing state-of-the-art HDA methods in terms of prediction accuracy and training efficiency. |
| first_indexed | 2024-10-01T03:41:25Z |
| format | Journal Article |
| id | ntu-10356/85717 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T03:41:25Z |
| publishDate | 2019 |
| record_format | dspace |
| spelling | ntu-10356/857172019-12-06T16:08:58Z Multi-class heterogeneous domain adaptation Zhou, Joey Tianyi Tsang, Ivor W. Pan, Sinno Jialin Tan, Mingkui School of Computer Science and Engineering Multi-class Classification Heterogeneous Domain Adaptation Engineering::Computer science and engineering A crucial issue in heterogeneous domain adaptation (HDA) is the ability to learn a feature mapping between different types of features across domains. Inspired by language translation, a word translated from one language corresponds to only a few words in another language, we present an efficient method named Sparse Heterogeneous Feature Representation (SHFR) in this paper for multi-class HDA to learn a sparse feature transformation between domains with multiple classes. Specifically, we formulate the problem of learning the feature transformation as a compressed sensing problem by building multiple binary classifiers in the target domain as various measurement sensors, which are decomposed from the target multi-class classification problem. We show that the estimation error of the learned transformation decreases with the increasing number of binary classifiers. In other words, for adaptation across heterogeneous domains to be successful, it is necessary to construct a sufficient number of incoherent binary classifiers from the original multi-class classification problem. To achieve this, we propose to apply the error correcting output correcting (ECOC) scheme to generate incoherent classifiers. To speed up the learning of the feature transformation across domains, we apply an efficient batch-mode algorithm to solve the resultant nonnegative sparse recovery problem. Theoretically, we present a generalization error bound of our proposed HDA method under a multi-class setting. Lastly, we conduct extensive experiments on both synthetic and real-world datasets to demonstrate the superiority of our proposed method over existing state-of-the-art HDA methods in terms of prediction accuracy and training efficiency. MOE (Min. of Education, S’pore) Published version 2019-07-10T08:18:37Z 2019-12-06T16:08:58Z 2019-07-10T08:18:37Z 2019-12-06T16:08:58Z 2019 Journal Article Zhou, J. T., Tsang, I. W., Pan, S. J., & Tan, M. (2019). Multi-class heterogeneous domain adaptation. Journal of Machine Learning Research, 20(57), 1-31. 1532-4435 https://hdl.handle.net/10356/85717 http://hdl.handle.net/10220/49259 http://jmlr.org/papers/v20/13-580.html en Journal of Machine Learning Research © 2019 Joey Tianyi Zhou, Ivor W. Tsang, Sinno Jialin Pan, Mingkui Tan. License: CC-BY 4.0, see https://creativecommons.org/licenses/by/4.0/. Attribution requirements are provided at http://jmlr.org/papers/v20/13-580.html. 31 p. application/pdf |
| spellingShingle | Multi-class Classification Heterogeneous Domain Adaptation Engineering::Computer science and engineering Zhou, Joey Tianyi Tsang, Ivor W. Pan, Sinno Jialin Tan, Mingkui Multi-class heterogeneous domain adaptation |
| title | Multi-class heterogeneous domain adaptation |
| title_full | Multi-class heterogeneous domain adaptation |
| title_fullStr | Multi-class heterogeneous domain adaptation |
| title_full_unstemmed | Multi-class heterogeneous domain adaptation |
| title_short | Multi-class heterogeneous domain adaptation |
| title_sort | multi class heterogeneous domain adaptation |
| topic | Multi-class Classification Heterogeneous Domain Adaptation Engineering::Computer science and engineering |
| url | https://hdl.handle.net/10356/85717 http://hdl.handle.net/10220/49259 http://jmlr.org/papers/v20/13-580.html |
| work_keys_str_mv | AT zhoujoeytianyi multiclassheterogeneousdomainadaptation AT tsangivorw multiclassheterogeneousdomainadaptation AT pansinnojialin multiclassheterogeneousdomainadaptation AT tanmingkui multiclassheterogeneousdomainadaptation |