Feature Selection Using a Neural Network With Group Lasso Regularization and Controlled Redundancy
Jian Wang, Huaqing Zhang, Junze Wang, Yi‐Fei Pu, Nikhil R. Pal
Abstract
-norm of weight matrix between the input and hidden layers. These penalty terms are nonsmooth at the origin, and hence, one simple but efficient smoothing technique is employed to overcome this issue. The monotonicity and convergence of the proposed algorithm are specified and proved under suitable assumptions. Then, extensive experiments are conducted on both artificial and real data sets. Empirical results explicitly demonstrate the ability of the proposed FS scheme and its effectiveness in controlling redundancy. The empirical simulations are observed to be consistent with the theoretical results.
Topics & Concepts
Feature selectionRedundancy (engineering)Regularization (linguistics)Artificial neural networkArtificial intelligenceComputer sciencePattern recognition (psychology)Lasso (programming language)Feature (linguistics)Machine learningOperating systemWorld Wide WebPhilosophyLinguisticsNeural Networks and ApplicationsFace and Expression RecognitionFault Detection and Control Systems