We consider the problem of variable group selection for least squares regression, namely, that of selecting groups of variables for best regression performance, leveraging and adhering to a natural grouping structure within the explanatory variables. We show that this problem can be efficiently addressed by using a certain greedy style algorithm. More precisely, we propose the Group Orthogonal ...

An automatic algorithm is derived for constructing kernel density estimates based on a regression approach that directly optimizes generalization capability. Computational efficiency of the density construction is ensured using an orthogonal forward regression, and the algorithm incrementally minimizes the leave-one-out test score. Local regularization is incorporated into the density construct...

A l-norm penalized orthogonal forward regression (l-POFR) algorithm is proposed based on the concept of leaveone-out mean square error (LOOMSE). Firstly, a new l-norm penalized cost function is defined in the constructed orthogonal space, and each orthogonal basis is associated with an individually tunable regularization parameter. Secondly, due to orthogonal computation, the LOOMSE can be anal...

The Lasso, the Forward Stagewise regression and the Lars are closely related procedures recently proposed for linear regression problems. Each of them can produce sparse models and can be used both for estimation and variable selection. In practical implementations these algorithms are typically tuned to achieve optimal prediction accuracy. We show that, when the prediction accuracy is used as ...

We consider the problem of variable group selection for least squares regression, namely, that of selecting groups of variables for best regression performance, leveraging and adhering to a natural grouping structure within the explanatory variables. We show that this problem can be efficiently addressed by using a certain greedy style algorithm. More precisely, we propose the Group Orthogonal ...

A novel iterative learning algorithm is proposed to improve the classic orthogonal forward regression (OFR) algorithm in an attempt to produce an optimal solution under a purely OFR framework without using any other auxiliary algorithms. The new algorithm searches for the optimal solution on a global solution space while maintaining the advantage of simplicity and computational efficiency. Both...

We develop an orthogonal forward selection (OFS) approach to construct radial basis function (RBF) network classifiers for two-class problems. Our approach integrates several concepts in probabilistic modelling, including cross validation, mutual information and Bayesian hyperparameter fitting. At each stage of the OFS procedure, one model term is selected by maximising the leave-one-out mutual...

In this thesis we propose the use of non-orthogonal forward link for future CDMA cellular systenis. The non-orthogonal forward lin& uses long PN codes as the spreading codes. The use of long PN codes has a few advantages over the common practice of using orthogonal codes. It ailows the design of a flexible air interface that can easily accommodate multimedia traffic, and has the potential to of...

In this paper we propose an efficient two-level model identification method for a large class of linear-in-the-parameters models from the observational data. A new elastic net orthogonal forward regression (ENOFR) algorithm is employed at the lower level to carry out simultaneous model selection and elastic net parameter estimation. The two regularization parameters in the elastic net are optim...