Sequential alternating proximal method for scalable sparse structural SVMs

Balamurugan, P and Shevade, Shirish and Babu, Ravindra T (2012) Sequential alternating proximal method for scalable sparse structural SVMs. In: 2012 IEEE 12th International Conference on Data Mining (ICDM), 10-13 Dec. 2012, Brussels, BELGIUM, pp. 61-70.

PDF Inte_Con_Dat_Min_61_2013.pdf - Published Version Restricted to Registered users only Download (359kB) | Request a copy

Abstract

Structural Support Vector Machines (SSVMs) have recently gained wide prominence in classifying structured and complex objects like parse-trees, image segments and Part-of-Speech (POS) tags. Typical learning algorithms used in training SSVMs result in model parameters which are vectors residing in a large-dimensional feature space. Such a high-dimensional model parameter vector contains many non-zero components which often lead to slow prediction and storage issues. Hence there is a need for sparse parameter vectors which contain a very small number of non-zero components. L1-regularizer and elastic net regularizer have been traditionally used to get sparse model parameters. Though L1-regularized structural SVMs have been studied in the past, the use of elastic net regularizer for structural SVMs has not been explored yet. In this work, we formulate the elastic net SSVM and propose a sequential alternating proximal algorithm to solve the dual formulation. We compare the proposed method with existing methods for L1-regularized Structural SVMs. Experiments on large-scale benchmark datasets show that the proposed dual elastic net SSVM trained using the sequential alternating proximal algorithm scales well and results in highly sparse model parameters while achieving a comparable generalization performance. Hence the proposed sequential alternating proximal algorithm is a competitive method to achieve sparse model parameters and a comparable generalization performance when elastic net regularized Structural SVMs are used on very large datasets.

Item Type:	Conference Paper
Series.:	IEEE International Conference on Data Mining
Publisher:	IEEE
Additional Information:	Copyright of this article belongs to IEEE.
Keywords:	Structural SVMs; Alternating Proximal Method
Department/Centre:	Division of Electrical Sciences > Computer Science & Automation
Date Deposited:	04 Jun 2013 11:59
Last Modified:	04 Jun 2013 11:59
URI:	http://eprints.iisc.ac.in/id/eprint/46531

Actions (login required)

View Item