A Kernel Method for the Optimization of the Margin Distribution

Authors:
Fabio Aiolli;Giovanni San Martino;Alessandro Sperduti
Affiliations:
Dept. of Pure and Applied Mathematics, , Padova, Italy 35131;Dept. of Pure and Applied Mathematics, , Padova, Italy 35131;Dept. of Pure and Applied Mathematics, , Padova, Italy 35131
Venue:
ICANN '08 Proceedings of the 18th international conference on Artificial Neural Networks, Part I
Year:
2008

Citing 6
Cited 1

Discrete neural computation: a theoretical foundation

Discrete neural computation: a theoretical foundation
Improved Generalization Through Explicit Optimization of Margins

Machine Learning
Soft Margins for AdaBoost

Machine Learning
A re-weighting strategy for improving margins

Artificial Intelligence
On generalization bounds, projection profile, and margin distribution

ICML '02 Proceedings of the Nineteenth International Conference on Machine Learning
How boosting the margin can also boost classifier complexity

ICML '06 Proceedings of the 23rd international conference on Machine learning

Margin distribution based bagging pruning

Neurocomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Recent results in theoretical machine learning seem to suggest that nice properties of the margin distribution over a training set turns out in a good performance of a classifier. The same principle has been already used in SVM and other kernel based methods as the associated optimization problems try to maximize the minimum of these margins.In this paper, we propose a kernel based method for the direct optimization of the margin distribution (KM-OMD). The method is motivated and analyzed from a game theoretical perspective. A quite efficient optimization algorithm is then proposed. Experimental results over a standard benchmark of 13 datasets have clearly shown state-of-the-art performances.