We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Linear classifiers are nearly optimal when hidden variables have diverse effects.
- Authors
Bshouty, Nader; Long, Philip
- Abstract
We analyze classification problems in which data is generated by a two-tiered random process. The class is generated first, then a layer of conditionally independent hidden variables, and finally the observed variables. For sources like this, the Bayes-optimal rule for predicting the class given the values of the observed variables is a two-layer neural network. We show that, if the hidden variables have non-negligible effects on many observed variables, a linear classifier approximates the error rate of the Bayes optimal classifier up to lower order terms. We also show that the hinge loss of a linear classifier is not much more than the Bayes error rate, which implies that an accurate linear classifier can be found efficiently.
- Subjects
STOCHASTIC processes; ARTIFICIAL neural networks; ERROR rates; MATHEMATICAL variables; SCIENTIFIC method
- Publication
Machine Learning, 2012, Vol 86, Issue 2, p209
- ISSN
0885-6125
- Publication type
Article
- DOI
10.1007/s10994-011-5262-7