MORE ABOUT THE DIFFERENCE BETWEEN MEN AND WOMEN - EVIDENCE FROM LINEAR NEURAL NETWORKS AND THE PRINCIPAL-COMPONENT APPROACH

The ability of a statistical/neural network to classify faces by sex b y means of a pixel-based representation has not been fully investigate d. Simulations with pixel-based codes have provided sex-classification results that are less impressive than those reported for measurement- based codes. In no case, however, have the reported pixel-based simula tions been optimized for the task of classifying faces by sex. A serie s of simulations is described in which four network models were applie d to the same pixel-based face code. These simulations involved either a radial basis function network or a perceptron as a classifier, prec eded or not by a preprocessing step of eigendecomposition. It is shown that performance comparable to that of the measurement-based models c an be achieved with pixel-based input (90%) when the data are preproce ssed. The effect of the eigendecomposition preprocessing of the faces is then compared with spatial-frequency analysis of face images and an alyzed in terms of the perceptual information it captures. It is shown that such an examination may offer insight into the facial aspects im portant to the sex-classification process. Finally, the contribution o f hair information to the performance of the model is evaluated. It is shown that, although the hair contributes to the sex-classification p rocess, it is not the only important contributor.