On the other hand, the L1-norm SVM has two drawbacks: (1) when there are several highly correlated variables, the L1-norm SVM tends to pick only a few of them, and remove the rest; (2) the number of ...

Lifeng Wang, Xiaotong Shen, On L1-Norm Multiclass Support Vector Machines: Methodology and Theory, Journal of the American Statistical Association, Vol. 102, No. 478 ...