Increasing Feature Selection Accuracy for L 1 Regularized Linear Models in Large Datasets

  • abhishek. jaiantilal
  • Published 2010

Abstract

L1 (also referred to as the 1-norm or Lasso) penalty based formulations have been shown to be effective in problem domains when noisy features are present. However, the L1 penalty does not give favorable asymptotic properties with respect to feature selection, and has been shown to be inconsistent as a feature selection estimator; e.g. when noisy features… (More)

6 Figures and Tables

Topics

  • Presentations referencing similar topics