Support Vector Learning Machines (SVM) are nding application in pattern recognition, regression estimation, and operator inversion for ill-posed problems. Against this very genera...
Abstract. Stacked generalization is a flexible method for multiple classifier combination; however, it tends to overfit unless the combiner function is sufficiently smooth. Prev...
The optimization method is one of key issues in discriminative learning of pattern classifiers. This paper proposes a hybrid approach of the Covariance Matrix Adaptation Evolution...
Abstract. We investigate a number of approaches to pose invariant face recognition. Basically, the methods involve three sequential functions for capturing nonlinear manifolds of f...
: In this work we present a strategy for off-line signature verification. It takes into account a writer-independent model which reduces the pattern recognition problem to a 2-clas...
Luiz S. Oliveira, Edson J. R. Justino, Robert Sabo...