Model Selection by Loss Rank for Classification and Unsupervised Learning
Methodology
2010-11-08 v1 Machine Learning
Abstract
Hutter (2007) recently introduced the loss rank principle (LoRP) as a generalpurpose principle for model selection. The LoRP enjoys many attractive properties and deserves further investigations. The LoRP has been well-studied for regression framework in Hutter and Tran (2010). In this paper, we study the LoRP for classification framework, and develop it further for model selection problems in unsupervised learning where the main interest is to describe the associations between input measurements, like cluster analysis or graphical modelling. Theoretical properties and simulation studies are presented.
Cite
@article{arxiv.1011.1379,
title = {Model Selection by Loss Rank for Classification and Unsupervised Learning},
author = {Minh-Ngoc Tran and Marcus Hutter},
journal= {arXiv preprint arXiv:1011.1379},
year = {2010}
}
Comments
20 pages, 2 figures