For high-dimensional data, most feature-selection methods, such as SIS and the lasso, involve ranking and selecting features individually. These methods do not require many computational resources, but they ignore feature interactions. A simple recursive approach, which, without requiring many...

Many contemporary classifiers are constructed to provide good performance for very high dimensional data. However, an issue that is at least as important as good classification is determining which of the many potential variables provide key information for good decisions. Responding to this...

In standard parametric classifiers, or classifiers based on nonparametric methods but where there is an opportunity for estimating population densities, the prior probabilities of the respective populations play a key role. However, those probabilities are largely ignored in the construction of...

The interpretation of generative, discriminative and hybrid approaches to classification is discussed, in particular for the generative-discriminative tradeoff (GDT), a hybrid approach. The asymptotic efficiency of the GDT, relative to that of its generative or discriminative counterpart, is...

The aims of this short note are two-fold. First, it shows that, for a random variable X, the area under the curve of its folded cumulative distribution function equals the mean absolute deviation (MAD) from the median. Such an equivalence implies that the MAD is the area between the cumulative...

This paper is concerned with the nonparametric estimation of regression quantiles where the response variable is randomly censored. Using results on the strong uniform convergence of U-processes, we derive a global Bahadur representation for the weighted local polynomial estimators, which is...