Column Subset Selection with Missing Data via Active Sampling

Yining Wang; Aarti Singh

Column Subset Selection with Missing Data via Active Sampling

Yining Wang, Aarti Singh ;

Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics, PMLR 38:1033-1041, 2015.

Abstract

Column subset selection of massive data matrices has found numerous applications in real-world data systems. In this paper, we propose and analyze two sampling based algorithms for column subset selection without access to the complete input matrix. To our knowledge, these are the first algorithms for column subset selection with missing data that are provably correct. The proposed methods work for row/column coherent matrices by employing the idea of adaptive sampling. Furthermore, when the input matrix has a noisy low-rank structure, one algorithm enjoys a relative error bound.

Related Material

@InProceedings{pmlr-v38-wang15c, title = {{Column Subset Selection with Missing Data via Active Sampling}}, author = {Yining Wang and Aarti Singh}, booktitle = {Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics}, pages = {1033--1041}, year = {2015}, editor = {Guy Lebanon and S. V. N. Vishwanathan}, volume = {38}, series = {Proceedings of Machine Learning Research}, address = {San Diego, California, USA}, month = {09--12 May}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v38/wang15c.pdf}, url = {http://proceedings.mlr.press/v38/wang15c.html}, abstract = {Column subset selection of massive data matrices has found numerous applications in real-world data systems. In this paper, we propose and analyze two sampling based algorithms for column subset selection without access to the complete input matrix. To our knowledge, these are the first algorithms for column subset selection with missing data that are provably correct. The proposed methods work for row/column coherent matrices by employing the idea of adaptive sampling. Furthermore, when the input matrix has a noisy low-rank structure, one algorithm enjoys a relative error bound.} }

%0 Conference Paper %T Column Subset Selection with Missing Data via Active Sampling %A Yining Wang %A Aarti Singh %B Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2015 %E Guy Lebanon %E S. V. N. Vishwanathan %F pmlr-v38-wang15c %I PMLR %J Proceedings of Machine Learning Research %P 1033--1041 %U http://proceedings.mlr.press %V 38 %W PMLR %X Column subset selection of massive data matrices has found numerous applications in real-world data systems. In this paper, we propose and analyze two sampling based algorithms for column subset selection without access to the complete input matrix. To our knowledge, these are the first algorithms for column subset selection with missing data that are provably correct. The proposed methods work for row/column coherent matrices by employing the idea of adaptive sampling. Furthermore, when the input matrix has a noisy low-rank structure, one algorithm enjoys a relative error bound.

TY - CPAPER TI - Column Subset Selection with Missing Data via Active Sampling AU - Yining Wang AU - Aarti Singh BT - Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics PY - 2015/02/21 DA - 2015/02/21 ED - Guy Lebanon ED - S. V. N. Vishwanathan ID - pmlr-v38-wang15c PB - PMLR SP - 1033 DP - PMLR EP - 1041 L1 - http://proceedings.mlr.press/v38/wang15c.pdf UR - http://proceedings.mlr.press/v38/wang15c.html AB - Column subset selection of massive data matrices has found numerous applications in real-world data systems. In this paper, we propose and analyze two sampling based algorithms for column subset selection without access to the complete input matrix. To our knowledge, these are the first algorithms for column subset selection with missing data that are provably correct. The proposed methods work for row/column coherent matrices by employing the idea of adaptive sampling. Furthermore, when the input matrix has a noisy low-rank structure, one algorithm enjoys a relative error bound. ER -

Wang, Y. & Singh, A.. (2015). Column Subset Selection with Missing Data via Active Sampling. Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics, in PMLR 38:1033-1041