A Note on k-support Norm Regularized Risk Minimization

Matthew Blaschko (1, 2)

1 GALEN - Organ Modeling through Extraction, Representation and Understanding of Medical Image Content
2 CVN - Centre de vision numérique

The k-support norm has been recently introduced to perform correlated sparsity regularization. Although Argyriou et al. only reported experiments using squared loss, here we apply it to several other commonly used settings resulting in novel machine learning algorithms with interesting and familiar limit cases. Source code for the algorithms described here is available.

Mots clés: k-support norm; structured sparsity; regularization; least-squares; hinge loss; support vector machine; SVM; regularized logistic regression; AdaBoost; support vector regression; SVR

Domaines: Apprentissage [cs.LG]

Matthew Blaschko. A Note on k-support Norm Regularized Risk Minimization. 2013. ⟨hal-00804592v2⟩