A new forward regression model identification algorithm is introduced. The derived model parameters, in each forward regression step, are initially estimated via orthogonal least squares (OLS) (using the modified Gram-Schmidt procedure), followed by being tuned with a new gradient descent learning algorithm based on the basis pursuit that minimizes the norm of the parameter estimate vector. The...

This work proposes a new relay selection algorithm in an opportunistic cooperative network, which aims to establish fairness among the users. Our approach provides the same overall outage probability for users at different locations. To this end, we first define a selection region containing Amplify-and-Forward (AF) relays with a superior channel quality. Then, opportunistic relay selection fro...

Using the classical Parzen window (PW) estimate as the target function, the sparse kernel density estimator is constructed in a forward constrained regression manner. The leave-one-out (LOO) test score is used for kernel selection. The jackknife parameter estimator subject to positivity constraint check is used for the parameter estimation of a single parameter at each forward step. As such the...

Feature selection methods, as a preprocessing step to machine learning, are effective in reducing dimensionality, removing irrelevant data, increasing learning accuracy, and improving result comprehensibility. However, the recent increase of dimensionality of data poses a severe challenge to many existing feature selection methods with respect to the efficiency and effectiveness. In this work, ...

We consider the problem of minimizing the differential delay in a virtually concatenated Ethernet over SONET (EoS) system by suitable path selection. In such a system, a service provider can dynamically add virtual channels to or drop them from a Virtually Concatenated Group (VCG). A new virtual channel can be added to the VCG provided that the differential delays between the new channel and th...

In spectrometric problems, objects are characterized by highresolution spectra that correspond to hundreds to thousands of variables. In this context, even fast variable selection methods lead to high computational load. However, spectra are generally smooth and can therefore be accurately approximated by splines. In this paper, we propose to use a B-spline expansion as a pre-processing step be...

In this study, the authors propose a new feature selection scheme, the incremental forward feature selection, which is inspired by incremental reduced support vector machines. In their method, a new feature is added into the current selected feature subset if it will bring in the most extra information. This information is measured by using the distance between the new feature vector and the co...