Skip to yearly menu bar Skip to main content


A Consistent Regularization Approach for Structured Prediction

Carlo Ciliberto · Lorenzo Rosasco · Alessandro Rudi

Area 5+6+7+8 #128

Keywords: [ Multi-task and Transfer Learning ] [ Learning Theory ] [ Kernel Methods ]


We propose and analyze a regularization approach for structured prediction problems. We characterize a large class of loss functions that allows to naturally embed structured outputs in a linear space. We exploit this fact to design learning algorithms using a surrogate loss approach and regularization techniques. We prove universal consistency and finite sample bounds characterizing the generalization properties of the proposed method. Experimental results are provided to demonstrate the practical usefulness of the proposed approach.

Live content is unavailable. Log in and register to view live content