Optimization Online


Stochastic Discrete First-order Algorithm for Feature Subset Selection

Kota Kudo(s1920464***at***s.tsukuba.ac.jp)
Yuichi Takano(ytakano***at***sk.tsukuba.ac.jp)
Ryo Nomura(nomu***at***waseda.jp)

Abstract: This paper addresses the problem of selecting a significant subset of candidate features to use for multiple linear regression. Bertsimas et al. (2016) recently proposed the discrete first-order (DFO) algorithm to efficiently find near-optimal solutions to this problem. However, this algorithm is unable to escape from locally optimal solutions. To resolve this, we propose a stochastic discrete first-order (SDFO) algorithm for feature subset selection. In this algorithm, random perturbations are added to a sequence of candidate solutions as a means to escape from locally optimal solutions, which broadens the range of discoverable solutions. Moreover, we derive the optimal step size in the gradient-descent direction to accelerate convergence of the algorithm. We also make effective use of the L2-regularization term to improve the predictive performance of a resultant subset regression model. The simulation results demonstrate that our algorithm substantially outperforms the original DFO algorithm. Our algorithm was superior in predictive performance to lasso and forward stepwise selection as well.

Keywords: feature subset selection, optimization algorithm, linear regression, machine learning, statistics

Category 1: Applications -- Science and Engineering (Statistics )

Category 2: Integer Programming


Download: [PDF]

Entry Submitted: 10/09/2019
Entry Accepted: 10/09/2019
Entry Last Modified: 10/09/2019

Modify/Update this entry

  Visitors Authors More about us Links
  Subscribe, Unsubscribe
Digest Archive
Search, Browse the Repository


Coordinator's Board
Classification Scheme
Give us feedback
Optimization Journals, Sites, Societies
Mathematical Optimization Society