TY - GEN
T1 - Gender classification of human faces using inference through contradictions
AU - Bai, Xue
AU - Cherkassky, Vladimir
PY - 2008/11/24
Y1 - 2008/11/24
N2 - We present an empirical study of gender classification of human faces, using new learning methodology called inference through contradictions, introduced in [9]. This approach allows to incorporate a priori knowledge in the form of additional (unlabeled) samples, called the Universum, into the supervised learning process. Application of this methodology to gender classification shows that using this approach enables better generalization over standard SVM classification (using labeled data alone).
AB - We present an empirical study of gender classification of human faces, using new learning methodology called inference through contradictions, introduced in [9]. This approach allows to incorporate a priori knowledge in the form of additional (unlabeled) samples, called the Universum, into the supervised learning process. Application of this methodology to gender classification shows that using this approach enables better generalization over standard SVM classification (using labeled data alone).
UR - http://www.scopus.com/inward/record.url?scp=56349103629&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=56349103629&partnerID=8YFLogxK
U2 - 10.1109/IJCNN.2008.4633879
DO - 10.1109/IJCNN.2008.4633879
M3 - Conference contribution
AN - SCOPUS:56349103629
SN - 9781424418213
T3 - Proceedings of the International Joint Conference on Neural Networks
SP - 746
EP - 750
BT - 2008 International Joint Conference on Neural Networks, IJCNN 2008
T2 - 2008 International Joint Conference on Neural Networks, IJCNN 2008
Y2 - 1 June 2008 through 8 June 2008
ER -