online read us now
Paper details
Number 4 - December 2014
Volume 24 - 2014
A primal sub-gradient method for structured classification with the averaged sum loss
Dejan Mančev, Branimir Todorović
Abstract
We present a primal sub-gradient method for structured SVM optimization defined with the averaged sum of hinge losses
inside each example. Compared with the mini-batch version of the Pegasos algorithm for the structured case, which deals
with a single structure from each of multiple examples, our algorithm considers multiple structures from a single example
in one update. This approach should increase the amount of information learned from the example. We show that the
proposed version with the averaged sum loss has at least the same guarantees in terms of the prediction loss as the stochastic
version. Experiments are conducted on two sequence labeling problems, shallow parsing and part-of-speech tagging, and
also include a comparison with other popular sequential structured learning algorithms.
Keywords
structured classification, support vector machines, sub-gradient methods, sequence labeling