TY - GEN
T1 - Learning using structured data
T2 - 2007 International Joint Conference on Neural Networks, IJCNN 2007
AU - Liang, Lichen
AU - Cherkassky, Vladimir
PY - 2007/12/1
Y1 - 2007/12/1
N2 - This paper investigates a new learning setting recently introduced by Vapnik [8] that takes into account a known structure of the training data to improve generalization performance. This setting is a special case of a new inference technology known as Learning with Hidden Information [8] suitable for many real-life applications with sparse high-dimensional data. We first briefly describe an extension of SVM called SVMγ+ [8] that is associated with this new learning setting, and verify its effectiveness using a synthetic data set. Then we demonstrate the effectiveness of SVMγ+ on a difficult real-life problem: detection of cognitive states from fMRI images obtained from different subjects. These empirical results show that the SVMγ+ approach achieves improved inter-subject generalization vs standard SVM technology.
AB - This paper investigates a new learning setting recently introduced by Vapnik [8] that takes into account a known structure of the training data to improve generalization performance. This setting is a special case of a new inference technology known as Learning with Hidden Information [8] suitable for many real-life applications with sparse high-dimensional data. We first briefly describe an extension of SVM called SVMγ+ [8] that is associated with this new learning setting, and verify its effectiveness using a synthetic data set. Then we demonstrate the effectiveness of SVMγ+ on a difficult real-life problem: detection of cognitive states from fMRI images obtained from different subjects. These empirical results show that the SVMγ+ approach achieves improved inter-subject generalization vs standard SVM technology.
UR - http://www.scopus.com/inward/record.url?scp=51749097342&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=51749097342&partnerID=8YFLogxK
U2 - 10.1109/IJCNN.2007.4371006
DO - 10.1109/IJCNN.2007.4371006
M3 - Conference contribution
AN - SCOPUS:51749097342
SN - 142441380X
SN - 9781424413805
T3 - IEEE International Conference on Neural Networks - Conference Proceedings
SP - 495
EP - 499
BT - The 2007 International Joint Conference on Neural Networks, IJCNN 2007 Conference Proceedings
Y2 - 12 August 2007 through 17 August 2007
ER -