An adaptive regularization method for sparse representation

Abstract

Sparse representation (SR) or sparse coding (SC), which assumes the data vector can be sparse represented by linear combination over basis vectors, has been successfully applied in machine learning and computer vision tasks. In order to solve sparse representation problem, regularization technique is applied to constrain the sparsity of coefficients of linear representation. In this paper, a reconstruction-error-based adaptive regularization parameter estimation method is proposed to improve the representation ability of SR. The adaptive regularization parameter aims to balance the reconstruction error and the sparsity of coefficient vector and to minimize reconstruction error. Substantial experiments are performed on some benchmark databases. Simulation results demonstrate that this adaptive regularization parameter estimation method can find a proper parameter for each test sample, consequently, can improve the accuracy of SR and eliminate a time-consuming cross-validation process.

Microscopic Machine Vision Based Degradation Monitoring of Low-Voltage Electromagnetic Coil Insulation Using Ensemble Learning in a Membrane Computing Framework

“…As well, more advanced machine learning methods, such as the 'generative adversarial networks' (GAN) [93] and graph theory based methods [94], can be included in the membrane computing framework for a more accurate matching result. In addition, more self-adaptive functions, like the nove global numerical optimization [95] and the adaptive regularization method [96], can be inserted into the membrane computing systems for a more rapid and robust matching result. Furthermore, more exquisite designs, as the electronic cluster eyes [97] and wireless sensors [98], are possible to make the system as a part of the internet of things.…”

A robust subspace classification scheme based on empirical intersection removal and sparse approximation

“…Although computationally simple, it is well known that this method yields a suboptimal solution to the regressor selection problem since all elements in w LS in general is non-zero. -Formulation using a sparse approximation To obtain a sparse solution, the approximation criterion need to include a term which promote a sparse solution [5,11,14,40]. If the least-squares criterion is augmented with an l 1 penalty on the weight vector…”