Heterogeneous Domain Adaptation for Multiple Classes

Joey Tianyi Zhou, Ivor W.Tsang, Sinno Jialin Pan, Mingkui Tan
Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics, PMLR 33:1095-1103, 2014.

Abstract

In this paper, we present an efficient Multi-class Heterogeneous Domain Adaptation (HDA) method, where data from the source and target domains are represented by heterogeneous features with different dimensions. Specifically, we propose to reconstruct a sparse feature transformation matrix to map the features of multiple classes from the source domain to the target domain. We cast this learning task as a compressed sensing problem, where each classifier can be deemed as a measurement sensor. Based on compressive sensing theory, the estimation error of the transformation matrix decreases with the increasing number of classifiers. Therefore, to guarantee the reconstruction performance, we construct sufficiently many binary classifiers based on the error correcting output correcting. Extensive experiments are conducted on both toy data and three real-world HDA applications to verify the superiority of our proposed method over existing state-of-the-art HDA methods in terms of prediction accuracy.

Cite this Paper

BibTeX
@InProceedings{pmlr-v33-zhou14, title = {{Heterogeneous Domain Adaptation for Multiple Classes}}, author = {Zhou, Joey Tianyi and W.Tsang, Ivor and Pan, Sinno Jialin and Tan, Mingkui}, booktitle = {Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics}, pages = {1095--1103}, year = {2014}, editor = {Kaski, Samuel and Corander, Jukka}, volume = {33}, series = {Proceedings of Machine Learning Research}, address = {Reykjavik, Iceland}, month = {22--25 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v33/zhou14.pdf}, url = {https://proceedings.mlr.press/v33/zhou14.html}, abstract = {In this paper, we present an efficient Multi-class Heterogeneous Domain Adaptation (HDA) method, where data from the source and target domains are represented by heterogeneous features with different dimensions. Specifically, we propose to reconstruct a sparse feature transformation matrix to map the features of multiple classes from the source domain to the target domain. We cast this learning task as a compressed sensing problem, where each classifier can be deemed as a measurement sensor. Based on compressive sensing theory, the estimation error of the transformation matrix decreases with the increasing number of classifiers. Therefore, to guarantee the reconstruction performance, we construct sufficiently many binary classifiers based on the error correcting output correcting. Extensive experiments are conducted on both toy data and three real-world HDA applications to verify the superiority of our proposed method over existing state-of-the-art HDA methods in terms of prediction accuracy.} }
Endnote
%0 Conference Paper %T Heterogeneous Domain Adaptation for Multiple Classes %A Joey Tianyi Zhou %A Ivor W.Tsang %A Sinno Jialin Pan %A Mingkui Tan %B Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2014 %E Samuel Kaski %E Jukka Corander %F pmlr-v33-zhou14 %I PMLR %P 1095--1103 %U https://proceedings.mlr.press/v33/zhou14.html %V 33 %X In this paper, we present an efficient Multi-class Heterogeneous Domain Adaptation (HDA) method, where data from the source and target domains are represented by heterogeneous features with different dimensions. Specifically, we propose to reconstruct a sparse feature transformation matrix to map the features of multiple classes from the source domain to the target domain. We cast this learning task as a compressed sensing problem, where each classifier can be deemed as a measurement sensor. Based on compressive sensing theory, the estimation error of the transformation matrix decreases with the increasing number of classifiers. Therefore, to guarantee the reconstruction performance, we construct sufficiently many binary classifiers based on the error correcting output correcting. Extensive experiments are conducted on both toy data and three real-world HDA applications to verify the superiority of our proposed method over existing state-of-the-art HDA methods in terms of prediction accuracy.
RIS
TY - CPAPER TI - Heterogeneous Domain Adaptation for Multiple Classes AU - Joey Tianyi Zhou AU - Ivor W.Tsang AU - Sinno Jialin Pan AU - Mingkui Tan BT - Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics DA - 2014/04/02 ED - Samuel Kaski ED - Jukka Corander ID - pmlr-v33-zhou14 PB - PMLR DP - Proceedings of Machine Learning Research VL - 33 SP - 1095 EP - 1103 L1 - http://proceedings.mlr.press/v33/zhou14.pdf UR - https://proceedings.mlr.press/v33/zhou14.html AB - In this paper, we present an efficient Multi-class Heterogeneous Domain Adaptation (HDA) method, where data from the source and target domains are represented by heterogeneous features with different dimensions. Specifically, we propose to reconstruct a sparse feature transformation matrix to map the features of multiple classes from the source domain to the target domain. We cast this learning task as a compressed sensing problem, where each classifier can be deemed as a measurement sensor. Based on compressive sensing theory, the estimation error of the transformation matrix decreases with the increasing number of classifiers. Therefore, to guarantee the reconstruction performance, we construct sufficiently many binary classifiers based on the error correcting output correcting. Extensive experiments are conducted on both toy data and three real-world HDA applications to verify the superiority of our proposed method over existing state-of-the-art HDA methods in terms of prediction accuracy. ER -
APA
Zhou, J.T., W.Tsang, I., Pan, S.J. & Tan, M.. (2014). Heterogeneous Domain Adaptation for Multiple Classes. Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 33:1095-1103 Available from https://proceedings.mlr.press/v33/zhou14.html.

Related Material