Poster
A Consistent Regularization Approach for Structured Prediction
Carlo Ciliberto · Lorenzo Rosasco · Alessandro Rudi
Area 5+6+7+8 #128
Keywords: [ Kernel Methods ] [ Learning Theory ] [ Multi-task and Transfer Learning ]
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.