Multi-label Classification via Feature-aware Implicit Label Space Encoding

Zijia Lin, Guiguang Ding, Mingqing Hu, Jianmin Wang
; Proceedings of the 31st International Conference on Machine Learning, PMLR 32(2):325-333, 2014.

Abstract

To tackle a multi-label classification problem with many classes, recently label space dimension reduction (LSDR) is proposed. It encodes the original label space to a low-dimensional latent space and uses a decoding process for recovery. In this paper, we propose a novel method termed FaIE to perform LSDR via Feature-aware Implicit label space Encoding. Unlike most previous work, the proposed FaIE makes no assumptions about the encoding process and directly learns a code matrix, i.e. the encoding result of some implicit encoding function, and a linear decoding matrix. To learn both matrices, FaIE jointly maximizes the recoverability of the original label space from the latent space, and the predictability of the latent space from the feature space, thus making itself feature-aware. FaIE can also be specified to learn an explicit encoding function, and extended with kernel tricks to handle non-linear correlations between the feature space and the latent space. Extensive experiments conducted on benchmark datasets well demonstrate its effectiveness.

Cite this Paper


BibTeX
@InProceedings{pmlr-v32-linc14, title = {Multi-label Classification via Feature-aware Implicit Label Space Encoding}, author = {Zijia Lin and Guiguang Ding and Mingqing Hu and Jianmin Wang}, booktitle = {Proceedings of the 31st International Conference on Machine Learning}, pages = {325--333}, year = {2014}, editor = {Eric P. Xing and Tony Jebara}, volume = {32}, number = {2}, series = {Proceedings of Machine Learning Research}, address = {Bejing, China}, month = {22--24 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v32/linc14.pdf}, url = {http://proceedings.mlr.press/v32/linc14.html}, abstract = {To tackle a multi-label classification problem with many classes, recently label space dimension reduction (LSDR) is proposed. It encodes the original label space to a low-dimensional latent space and uses a decoding process for recovery. In this paper, we propose a novel method termed FaIE to perform LSDR via Feature-aware Implicit label space Encoding. Unlike most previous work, the proposed FaIE makes no assumptions about the encoding process and directly learns a code matrix, i.e. the encoding result of some implicit encoding function, and a linear decoding matrix. To learn both matrices, FaIE jointly maximizes the recoverability of the original label space from the latent space, and the predictability of the latent space from the feature space, thus making itself feature-aware. FaIE can also be specified to learn an explicit encoding function, and extended with kernel tricks to handle non-linear correlations between the feature space and the latent space. Extensive experiments conducted on benchmark datasets well demonstrate its effectiveness.} }
Endnote
%0 Conference Paper %T Multi-label Classification via Feature-aware Implicit Label Space Encoding %A Zijia Lin %A Guiguang Ding %A Mingqing Hu %A Jianmin Wang %B Proceedings of the 31st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2014 %E Eric P. Xing %E Tony Jebara %F pmlr-v32-linc14 %I PMLR %J Proceedings of Machine Learning Research %P 325--333 %U http://proceedings.mlr.press %V 32 %N 2 %W PMLR %X To tackle a multi-label classification problem with many classes, recently label space dimension reduction (LSDR) is proposed. It encodes the original label space to a low-dimensional latent space and uses a decoding process for recovery. In this paper, we propose a novel method termed FaIE to perform LSDR via Feature-aware Implicit label space Encoding. Unlike most previous work, the proposed FaIE makes no assumptions about the encoding process and directly learns a code matrix, i.e. the encoding result of some implicit encoding function, and a linear decoding matrix. To learn both matrices, FaIE jointly maximizes the recoverability of the original label space from the latent space, and the predictability of the latent space from the feature space, thus making itself feature-aware. FaIE can also be specified to learn an explicit encoding function, and extended with kernel tricks to handle non-linear correlations between the feature space and the latent space. Extensive experiments conducted on benchmark datasets well demonstrate its effectiveness.
RIS
TY - CPAPER TI - Multi-label Classification via Feature-aware Implicit Label Space Encoding AU - Zijia Lin AU - Guiguang Ding AU - Mingqing Hu AU - Jianmin Wang BT - Proceedings of the 31st International Conference on Machine Learning PY - 2014/01/27 DA - 2014/01/27 ED - Eric P. Xing ED - Tony Jebara ID - pmlr-v32-linc14 PB - PMLR SP - 325 DP - PMLR EP - 333 L1 - http://proceedings.mlr.press/v32/linc14.pdf UR - http://proceedings.mlr.press/v32/linc14.html AB - To tackle a multi-label classification problem with many classes, recently label space dimension reduction (LSDR) is proposed. It encodes the original label space to a low-dimensional latent space and uses a decoding process for recovery. In this paper, we propose a novel method termed FaIE to perform LSDR via Feature-aware Implicit label space Encoding. Unlike most previous work, the proposed FaIE makes no assumptions about the encoding process and directly learns a code matrix, i.e. the encoding result of some implicit encoding function, and a linear decoding matrix. To learn both matrices, FaIE jointly maximizes the recoverability of the original label space from the latent space, and the predictability of the latent space from the feature space, thus making itself feature-aware. FaIE can also be specified to learn an explicit encoding function, and extended with kernel tricks to handle non-linear correlations between the feature space and the latent space. Extensive experiments conducted on benchmark datasets well demonstrate its effectiveness. ER -
APA
Lin, Z., Ding, G., Hu, M. & Wang, J.. (2014). Multi-label Classification via Feature-aware Implicit Label Space Encoding. Proceedings of the 31st International Conference on Machine Learning, in PMLR 32(2):325-333

Related Material