Learning iteratively a classifier with the Bayesian Model Averaging Principlestar
Rudy Sicard, Thierry Artières and Eric Petit
Pattern Recognition Volume 41, Number 3, pp. 930-938, 2008.

This is the latest version of this eprint.

Abstract

We present a learning algorithm for nominal vector data. It builds a complex classifier by adding iteratively a simple function that modifies the current classifier. In order to limit overtraining problem we focus on a class of such functions for which optimal Bayesian learning is tractable. We investigate a few classes of functions that yield to models that are similar to Naı¨ve Bayes and logistic classification. We report experimental results for a collection of standard data sets that show that our learning algorithm outperforms standard learning of such these standard models.

Available Versions of this Item

• Patch Learning for Incremental Classifier Design (deposited 22 November 2006)
  • Learning iteratively a classifier with the Bayesian Model Averaging Principlestar (deposited 14 February 2008) [Currently Displayed]

[Edit]