Creator:

Mančev, Dejan ; Todorović, Branimir

Contributor:

Korbicz, Józef (1951- ) - red. ; Uciński, Dariusz - red.

Title:

A primal sub-gradient method for structured classification with the averaged sum loss

Group publication title:

AMCS, Volume 24 (2014)

Subject and Keywords:

structured classification ; support vector machines ; sub-gradient methods ; sequence labeling

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.

Publisher:

Zielona Góra: Uniwersytet Zielonogórski

Date:

2014

Resource Type:

artykuł

DOI:

10.2478/amcs-2014-0067

Pages:

917-930

Source:

AMCS, volume 24, number 4 (2014) ; click here to follow the link

Language:

eng

License CC BY 4.0:

click here to follow the link

Rights:

Biblioteka Uniwersytetu Zielonogórskiego