Resumen
Method for eliminating bias (by age, ethnicity or gender) in biometric recognition systems, which involves defining a set of M samples of Y different people labeled based on attributes such as gender, ethnicity or age, where samples A and samples P correspond to samples of the same identity, while samples N correspond to different identities and where, in addition, a value corresponding to the bias of each sample is also entered; and where the proposed method is characterized in that it comprises the steps necessary to learn a transformation function φ (X) that generates a new feature space that allows: (i) to minimize distance d (XA, .XP) between feature vectors (XA, .XP) of A and P; (ii) maximize distance d (XA, .XN) between feature vectors (XA, .XN) A and N; and (iii) reduce bias or in samples until elimination, thus guaranteeing unbiased decision making. (Machine-translation by Google Translate, not legally binding)