TY  - JOUR
N2  - The Column Subset Selection Problem is a hard combinatorial optimization problem that provides a natural framework for unsupervised feature selection, and there exist efficient algorithms that provide good approximations. The drawback of the problem formulation is that it incorporates no form of regularization, and is therefore very sensitive to noise when presented with scarce data. In this paper we propose a regularized formulation of this problem, and derive a correct greedy algorithm that is similar in efficiency to existing greedy methods for the unregularized problem. We study its adequacy for feature selection and propose suitable formulations. Additionally, we derive a lower bound for the error of the proposed problems. Through various numerical experiments on real and synthetic data, we demonstrate the significantly increased robustness and stability of our method, as well as the improved conditioning of its output, all while remaining efficient for practical use.
EP  - 418
TI  - Regularized greedy column subset selection
VL  - 486
ID  - upm64356
JF  - Information Sciences
AV  - public
PB  - Elsevier
Y1  - 2019/06//
A1  - Ordozgoiti Rubio, Bruno
A1  - Mozo Velasco, Bonifacio Alberto
A1  - Garcia Lopez De Lacalle, Jesus
UR  - https://www.sciencedirect.com/science/article/abs/pii/S0020025519301495
SN  - 0020-0255
SP  - 393
KW  - Feature selection; Column subset selection; Unsupervised learning
ER  -