By Topic

Why error measures are sub-optimal for training neural network pattern classifiers

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

2 Author(s)
Hampshire, J.B., II ; Dept. of Electr. & Comput. Eng., Carnegie Mellon Univ., Pittsburgh, PA, USA ; Kumar, B.V.K.V.

Pattern classifiers that are trained in a supervised fashion are typically trained with an error measure objective function such as mean-squared error (MSE) or cross-entropy (CE). These classifiers can in theory yield Bayesian discrimination, but in practice they often fail to do so. The authors explain why this happens and identify a number of characteristics that the optimal objective function for training classifiers must have. They show that classification figures of merit (CFMmono) possess these optimal characteristics, whereas error measures such as MSE and CE do not. The arguments are illustrated with a simple example in which a CFMmono-trained low-order polynomial neural network approximates Bayesian discrimination on a random scalar with the fewest number of training samples and the minimum functional complexity necessary for the task. A comparable MSE-trained net yields significantly worse discrimination on the same task

Published in:

Neural Networks, 1992. IJCNN., International Joint Conference on  (Volume:4 )

Date of Conference:

7-11 Jun 1992