Improvement of Restricted Boltzmann Machine by Sparse Representation Based on Lorentz Function
Libin Chen, Weibao Zou · 2018
Restricted Boltzmann machine (RBM) is an effective feature extraction algorithm. Inspired by the visual cortex sparse representation, sparsity is introduced into RBM to get sparse representation. In the paper, the Lorentz function is introduced to RBM to construct constrained sparse RBM model (LRBM) to improve RBM. The experimental results show that the LRBM model can effectively extract the feature information. The average classification rate increases by 2% than that of RBM.