Minimization of for Compressed Sensing

P Yin, Y Lou, Q He, J Xin - SIAM Journal on Scientific Computing, 2015 - SIAM
We study minimization of the difference of \ell_1 and \ell_2 norms as a nonconvex and
Lipschitz continuous metric for solving constrained and unconstrained compressed sensing …

Deephoyer: Learning sparser neural network with differentiable scale-invariant sparsity measures

H Yang, W Wen, H Li - arXiv preprint arXiv:1908.09979, 2019 - arxiv.org
In seeking for sparse and efficient neural network models, many previous works investigated
on enforcing L1 or L0 regularizers to encourage weight sparsity during training. The L0 …

A weighted difference of anisotropic and isotropic total variation model for image processing

Y Lou, T Zeng, S Osher, J Xin - SIAM Journal on Imaging Sciences, 2015 - SIAM
We propose a weighted difference of anisotropic and isotropic total variation (TV) as a
regularization for image processing tasks, based on the well-known TV model and natural …

Computing Sparse Representation in a Highly Coherent Dictionary Based on Difference of and

Y Lou, P Yin, Q He, J Xin - Journal of Scientific Computing, 2015 - Springer
We study analytical and numerical properties of the L_1-L_2 L 1-L 2 minimization problem
for sparse representation of a signal over a highly coherent dictionary. Though the L_1-L_2 …

A scale-invariant approach for sparse signal recovery

Y Rahimi, C Wang, H Dong, Y Lou - SIAM Journal on Scientific Computing, 2019 - SIAM
In this paper, we study the ratio of the L_1 and L_2 norms, denoted as L_1/L_2, to promote
sparsity. Due to the nonconvexity and nonlinearity, there has been little attention to this scale …

Limited-Angle CT Reconstruction via the Minimization

C Wang, M Tao, JG Nagy, Y Lou - SIAM Journal on Imaging Sciences, 2021 - SIAM
In this paper, we consider minimizing the L_1/L_2 term on the gradient for a limited-angle
scanning problem in computed tomography (CT) reconstruction. We design a specific …

Accelerated Schemes for the Minimization

C Wang, M Yan, Y Rahimi, Y Lou - IEEE Transactions on Signal …, 2020 - ieeexplore.ieee.org
In this paper, we consider the L 1/L 2 minimization for sparse recovery and study its
relationship with the L 1-αL 2 model. Based on this relationship, we propose three numerical …

Hyperspectral image denoising via global spatial-spectral total variation regularized nonconvex local low-rank tensor approximation

H Zeng, X Xie, J Ning - Signal processing, 2021 - Elsevier
Hyperspectral image (HSI) denoising aims to restore clean HSI from the noise-contaminated
one which is usually caused during data acquisition and conversion. In this paper, we …

ℓ1− αℓ2 minimization methods for signal and image reconstruction with impulsive noise removal

P Li, W Chen, H Ge, MK Ng - Inverse Problems, 2020 - iopscience.iop.org
In this paper, we study ℓ 1− αℓ 2 (0< α⩽ 1) minimization methods for signal and image
reconstruction with impulsive noise removal. The data fitting term is based on ℓ 1 fidelity …

Sparse, efficient, and semantic mixture invariant training: Taming in-the-wild unsupervised sound separation

S Wisdom, A Jansen, RJ Weiss… - … IEEE Workshop on …, 2021 - ieeexplore.ieee.org
Supervised neural network training has led to significant progress on single-channel sound
separation. This approach relies on ground truth isolated sources, which precludes scaling …