Rademacher complexity and structural risk minimization: an application to human gene expression datasets

Authors:
Luca Oneto;Davide Anguita;Alessandro Ghio;Sandro Ridella
Affiliations:
DITEN, University of Genova, Genova, Italy;DITEN, University of Genova, Genova, Italy;DITEN, University of Genova, Genova, Italy;DITEN, University of Genova, Genova, Italy
Venue:
ICANN'12 Proceedings of the 22nd international conference on Artificial Neural Networks and Machine Learning - Volume Part II
Year:
2012

Citing 8
Cited 0

Model Selection and Error Estimation

Machine Learning
Quantum optimization for training support vector machines

Neural Networks - 2003 Special issue: Advances in neural networks research — IJCNN'03
Kernel Methods for Pattern Analysis

Kernel Methods for Pattern Analysis
Is cross-validation valid for small-sample microarray classification?

Bioinformatics
A comprehensive evaluation of multicategory classification methods for microarray gene expression cancer diagnosis

Bioinformatics
Trading convexity for scalability

ICML '06 Proceedings of the 23rd international conference on Machine learning
A study of cross-validation and bootstrap for accuracy estimation and model selection

IJCAI'95 Proceedings of the 14th international joint conference on Artificial intelligence - Volume 2
Maximal Discrepancy for Support Vector Machines

Neurocomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we target the problem of model selection for Support Vector Classifiers through in–sample methods, which are particularly appealing in the small–sample regime, i.e. when few high–dimensional patterns are available. In particular, we describe the application of a trimmed hinge loss function to Rademacher Complexity and Maximal Discrepancy based in–sample approaches. We also show that the selected classifiers outperform the ones obtained with other state-of-the-art in-sample and out–of–sample model selection techniques in classifying Human Gene Expression datasets.