LGMar 26, 2013
A Note on k-support Norm Regularized Risk Minimization
arXiv:1303.6390v25 citations
Originality Synthesis-oriented
AI Analysis
This work is incremental, extending an existing regularization method to more settings in machine learning.
The paper applied the k-support norm to various loss functions beyond squared loss, developing new machine learning algorithms with familiar limit cases.
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.