The Bin Model

Abu-Mostafa, Yaser and Song, Xubo and Nicholson, Alexander and Magdon-Ismail, Malik (2004) The Bin Model. Technical Report. California Institute of Technology, Pasadena, CA. [CaltechCSTR:2004.002]

Full text available as:

	Postscript - Requires a viewer, such as GhostView
	PDF - Requires Adobe Acrobat Reader or other PDF viewer.

Abstract

We propose a novel theoretical framework for understanding learning and generalization which we will call the bin model. Using the bin model, a closed form is derived for the generalization error that estimates the out-of-sample performance in terms of the in-sample performance. We address the problem of overfitting, and show that using a simple exhaustive learning algorithm it does not arise. This is independent of the target function, input distribution and learning model, and remains true even with noisy data sets. We apply our analysis to both classification and regression problems and give an example of how it may be used effectively in practice.

EPrint Type:	Monograph (Technical Report)
Uncontrolled Keywords:	Learning theory, generalization, noisy data, VC dimension, out of sample error, overfitting
Subjects:	All Records
ID Code:	525
Deposited By:	Caltech Library System
Deposited On:	06 July 2004
Record Number:	CaltechCSTR:2004.002
Official Persistent URL:	http://resolver.caltech.edu/CaltechCSTR:2004.002
Usage Policy:	You are granted permission for individual, educational, research and non-commercial reproduction, distribution, display and performance of this work in any format.

Archive Staff Only: edit this record