Canonical Correlation Analysis based on Hilbert-Schmidt Independence Criterion and Centered Kernel Target Alignment

Billy Chang, Uwe Kruger, Rafal Kustra, Junping Zhang
Proceedings of the 30th International Conference on Machine Learning, PMLR 28(2):316-324, 2013.

Abstract

Canonical correlation analysis (CCA) is a well established technique for identifying linear relationships among two variable sets. Kernel CCA (KCCA) is the most notable nonlinear extension but it lacks interpretability and robustness against irrelevant features. The aim of this article is to introduce two nonlinear CCA extensions that rely on the recently proposed Hilbert-Schmidt independence criterion and the centered kernel target alignment. These extensions determine linear projections that provide maximally dependent projected data pairs. The paper demonstrates that the use of linear projections allows removing irrelevant features, whilst extracting combinations of strongly associated features. This is exemplified through a simulation and the analysis of recorded data that are available in the literature.

Cite this Paper


BibTeX
@InProceedings{pmlr-v28-chang13, title = {Canonical Correlation Analysis based on Hilbert-Schmidt Independence Criterion and Centered Kernel Target Alignment}, author = {Chang, Billy and Kruger, Uwe and Kustra, Rafal and Zhang, Junping}, booktitle = {Proceedings of the 30th International Conference on Machine Learning}, pages = {316--324}, year = {2013}, editor = {Dasgupta, Sanjoy and McAllester, David}, volume = {28}, number = {2}, series = {Proceedings of Machine Learning Research}, address = {Atlanta, Georgia, USA}, month = {17--19 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v28/chang13.pdf}, url = {https://proceedings.mlr.press/v28/chang13.html}, abstract = {Canonical correlation analysis (CCA) is a well established technique for identifying linear relationships among two variable sets. Kernel CCA (KCCA) is the most notable nonlinear extension but it lacks interpretability and robustness against irrelevant features. The aim of this article is to introduce two nonlinear CCA extensions that rely on the recently proposed Hilbert-Schmidt independence criterion and the centered kernel target alignment. These extensions determine linear projections that provide maximally dependent projected data pairs. The paper demonstrates that the use of linear projections allows removing irrelevant features, whilst extracting combinations of strongly associated features. This is exemplified through a simulation and the analysis of recorded data that are available in the literature.} }
Endnote
%0 Conference Paper %T Canonical Correlation Analysis based on Hilbert-Schmidt Independence Criterion and Centered Kernel Target Alignment %A Billy Chang %A Uwe Kruger %A Rafal Kustra %A Junping Zhang %B Proceedings of the 30th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2013 %E Sanjoy Dasgupta %E David McAllester %F pmlr-v28-chang13 %I PMLR %P 316--324 %U https://proceedings.mlr.press/v28/chang13.html %V 28 %N 2 %X Canonical correlation analysis (CCA) is a well established technique for identifying linear relationships among two variable sets. Kernel CCA (KCCA) is the most notable nonlinear extension but it lacks interpretability and robustness against irrelevant features. The aim of this article is to introduce two nonlinear CCA extensions that rely on the recently proposed Hilbert-Schmidt independence criterion and the centered kernel target alignment. These extensions determine linear projections that provide maximally dependent projected data pairs. The paper demonstrates that the use of linear projections allows removing irrelevant features, whilst extracting combinations of strongly associated features. This is exemplified through a simulation and the analysis of recorded data that are available in the literature.
RIS
TY - CPAPER TI - Canonical Correlation Analysis based on Hilbert-Schmidt Independence Criterion and Centered Kernel Target Alignment AU - Billy Chang AU - Uwe Kruger AU - Rafal Kustra AU - Junping Zhang BT - Proceedings of the 30th International Conference on Machine Learning DA - 2013/05/13 ED - Sanjoy Dasgupta ED - David McAllester ID - pmlr-v28-chang13 PB - PMLR DP - Proceedings of Machine Learning Research VL - 28 IS - 2 SP - 316 EP - 324 L1 - http://proceedings.mlr.press/v28/chang13.pdf UR - https://proceedings.mlr.press/v28/chang13.html AB - Canonical correlation analysis (CCA) is a well established technique for identifying linear relationships among two variable sets. Kernel CCA (KCCA) is the most notable nonlinear extension but it lacks interpretability and robustness against irrelevant features. The aim of this article is to introduce two nonlinear CCA extensions that rely on the recently proposed Hilbert-Schmidt independence criterion and the centered kernel target alignment. These extensions determine linear projections that provide maximally dependent projected data pairs. The paper demonstrates that the use of linear projections allows removing irrelevant features, whilst extracting combinations of strongly associated features. This is exemplified through a simulation and the analysis of recorded data that are available in the literature. ER -
APA
Chang, B., Kruger, U., Kustra, R. & Zhang, J.. (2013). Canonical Correlation Analysis based on Hilbert-Schmidt Independence Criterion and Centered Kernel Target Alignment. Proceedings of the 30th International Conference on Machine Learning, in Proceedings of Machine Learning Research 28(2):316-324 Available from https://proceedings.mlr.press/v28/chang13.html.

Related Material