TY - GEN

T1 - Simultaneous learning of several Bayesian and Mahalanobis discriminant functions by a neural network with additional nodes

AU - Ito, Yoshifusa

AU - Izumi, Hiroyuki

AU - Srinivasan, Cidambi

PY - 2011

Y1 - 2011

N2 - We construct a neural network which can simultaneously approximate several Bayesian and Mahalanobis discriminant functions. The main part of the network is an ordinary one-hidden-layer neural network with a nonlinear output unit, but it has several additional nodes. Since the network has a task to approximate Mahalanobis discriminant functions, the state-conditional probability distributions are supposed to be normal distributions. The method is useful when the Bayesian discriminant functions can be decomposed into sums of a common main part and individual linear additional parts. The main part of the network approximates the quadratic part of the discriminant functions.

AB - We construct a neural network which can simultaneously approximate several Bayesian and Mahalanobis discriminant functions. The main part of the network is an ordinary one-hidden-layer neural network with a nonlinear output unit, but it has several additional nodes. Since the network has a task to approximate Mahalanobis discriminant functions, the state-conditional probability distributions are supposed to be normal distributions. The method is useful when the Bayesian discriminant functions can be decomposed into sums of a common main part and individual linear additional parts. The main part of the network approximates the quadratic part of the discriminant functions.

UR - http://www.scopus.com/inward/record.url?scp=80054723806&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80054723806&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.2011.6033294

DO - 10.1109/IJCNN.2011.6033294

M3 - Conference contribution

AN - SCOPUS:80054723806

SN - 9781457710865

T3 - Proceedings of the International Joint Conference on Neural Networks

SP - 733

EP - 740

BT - 2011 International Joint Conference on Neural Networks, IJCNN 2011 - Final Program

Y2 - 31 July 2011 through 5 August 2011

ER -