000185082 001__ 185082
000185082 005__ 20190316235615.0
000185082 0247_ $$2doi$$a10.1109/Cvpr.2013.259
000185082 02470 $$2ISI$$a000331094302007
000185082 037__ $$aCONF
000185082 245__ $$aLearning for Structured Prediction Using Approximate Subgradient Descent with Working Sets
000185082 269__ $$a2013
000185082 260__ $$c2013
000185082 336__ $$aConference Papers
000185082 520__ $$aWe propose a working set based approximate subgradient descent algorithm to minimize the margin-sensitive hinge loss arising from the soft constraints in max-margin learning frameworks, such as the structured SVM. We focus on the setting of general graphical models, such as loopy MRFs and CRFs commonly used in image segmentation, where exact inference is intractable and the most violated constraints can only be approximated, voiding the optimality guarantees of the structured SVM's cutting plane algorithm as well as reducing the robustness of existing subgradient based methods. We show that the proposed method obtains better approximate subgradients through the use of working sets, leading to improved convergence properties and increased reliability. Furthermore, our method allows new constraints to be randomly sampled instead of computed using the more expensive approximate inference techniques such as belief propagation and graph cuts, which can be used to reduce learning time at only a small cost of performance. We demonstrate the strength of our method empirically on the segmentation of a new publicly available electron microscopy dataset as well as the popular MSRC data set and show state-of-the-art results.
000185082 6531_ $$aStructured prediction
000185082 6531_ $$aMachine learning
000185082 6531_ $$aImage segmentation
000185082 700__ $$0242715$$aLucchi, Aurélien$$g185205
000185082 700__ $$0244419$$aLi, Yunpeng$$g201504
000185082 700__ $$0240252$$aFua, Pascal$$g112366
000185082 7112_ $$aConference on Computer Vision and Pattern Recognition (CVPR)$$cPortland, Oregon, USA$$dJune 23-28, 2013
000185082 8564_ $$s163365$$uhttps://infoscience.epfl.ch/record/185082/files/supplementary.pdf$$ySupplementary material$$zSupplementary material
000185082 8564_ $$s3750083$$uhttps://infoscience.epfl.ch/record/185082/files/top.pdf$$yn/a$$zn/a
000185082 909C0 $$0252087$$pCVLAB$$xU10659
000185082 909CO $$ooai:infoscience.tind.io:185082$$pconf$$pIC$$qGLOBAL_SET
000185082 917Z8 $$x185205
000185082 917Z8 $$x185205
000185082 917Z8 $$x185205
000185082 917Z8 $$x185205
000185082 917Z8 $$x112366
000185082 937__ $$aEPFL-CONF-185082
000185082 973__ $$aEPFL$$rREVIEWED$$sACCEPTED
000185082 980__ $$aCONF