This paper presents a new subband adaptive filter (SAF) algorithm for system identification scenario under impulsive interference, named generalized continuous mixed p-norm SAF (GCMPN-SAF) algorithm. The proposed uses GCMPN cost function to combat the interference. To further accelerate convergence rate in sparse and block-sparse processes, proportionate versions of algorithm, L0-norm GCMPN-SAF...

This deals with the problem of recovering a high-resolution digital image from one low resolution digital image and proposes a super-resolution algorithm based on the mixed l0/l1 norm minimization. Introducing some assumptions and focusing the uniformity and the gradation of the image, this paper formulates the colorization problem as a mixed l0/l1 norm minimization and proposes the algorithm b...

Recently, there has been focus on penalized loglikelihood covariance estimation for sparse inverse covariance (precision) matrices. The penalty is responsible for inducing sparsity, and a very common choice is the convex l1 norm. However, the best estimator performance is not always achieved with this penalty. The most natural sparsity promoting “norm” is the non-convex l0 penalty but its lack ...

We propose a practical method for L0 norm regularization for neural networks: pruning the network during training by encouraging weights to become exactly zero. Such regularization is interesting since (1) it can greatly speed up training and inference, and (2) it can improve generalization. AIC and BIC, well-known model selection criteria, are special cases of L0 regularization. However, since...

Numerical algorithms for the l0-norm regularized non-smooth non-convex minimization problems have recently became a topic of great interest within signal processing, compressive sensing, statistics, and machine learning. Nevertheless, the l0norm makes the problem combinatorial and generally computationally intractable. In this paper, we construct a new surrogate function to approximate l0-norm ...

This paper proposes an improved least mean kurtosis (LMK) algorithm based on l0-norm cost for enhancing the filter performance in a sparse system. The LMK adaptive filtering algorithm uses a kurtosis of an estimated error signal to improve the filter performance when the noise contamination is serious. Due to the influence of l0-norm cost, the proposed LMK algorithm ensures a fast convergence r...

To exploit the sparsity of the considered system, the diffusion proportionate-type least mean square (PtLMS) algorithms assign different gains to each tap in the convergence stage while the diffusion sparsity-constrained LMS (ScLMS) algorithms pull the components towards zeros in the steady-state stage. In this paper, by minimizing a differentiable cost function that utilizes the Riemannian dis...

Sparse representation has been applied successfully in many image analysis applications, including abnormal event detection, in which a baseline is to learn a dictionary from the training data and detect anomalies from its sparse codes. During this procedure, sparse codes which can be achieved through finding the L0-norm solution of the problem: min ‖Y −Dα‖2 +‖α‖0, is crucial. Note that D refer...

Non-negative matrix factorization (NMF), i.e. V ≈ WH where both V, W and H are non-negative has become a widely used blind source separation technique due to its part based representation. The NMF decomposition is not in general unique and a part based representation not guaranteed. However, imposing sparseness both improves the uniqueness of the decomposition and favors part based representati...

In this paper, we propose a formulation of a feature selecting support vector machine based on the L0-norm. We explore a perspective relaxation of the optimization problem and solve it using mixed-integer nonlinear programming (MINLP) techniques. Given a training set of labeled data instances, we construct a maxmargin classifier that minimizes the hinge loss as well as the cardinality of the we...