Insensitive stochastic gradient twin support vector machines for large scale problems

DOI10.1016/J.INS.2018.06.007zbMATH Open1440.68245DBLPjournals/isci/WangSBLLD18arXiv1704.05596OpenAlexW2805876623WikidataQ59305688 ScholiaQ59305688MaRDI QIDQ2198233FDOQ2198233

Authors: Zhen Wang, Yuan-Hai Shao, Lan Bai, Chunna Li, Liming Liu, Nai-Yang Deng 

Publication date: 9 September 2020

Published in: Information Sciences (Search for Journal in Brave)

Abstract: Stochastic gradient descent algorithm has been successfully applied on support vector machines (called PEGASOS) for many classification problems. In this paper, stochastic gradient descent algorithm is investigated to twin support vector machines for classification. Compared with PEGASOS, the proposed stochastic gradient twin support vector machines (SGTSVM) is insensitive on stochastic sampling for stochastic gradient descent algorithm. In theory, we prove the convergence of SGTSVM instead of almost sure convergence of PEGASOS. For uniformly sampling, the approximation between SGTSVM and twin support vector machines is also given, while PEGASOS only has an opportunity to obtain an approximation of support vector machines. In addition, the nonlinear SGTSVM is derived directly from its linear case. Experimental results on both artificial datasets and large scale problems show the stable performance of SGTSVM with a fast learning speed.

Full work available at URL: https://arxiv.org/abs/1704.05596

Recommendations

• A novel twin parametric support vector machine for large scale problem
• Pegasos: primal estimated sub-gradient solver for SVM
• Large-scale linear nonparallel support vector machine solver
• A coordinate descent margin based-twin support vector machine for classification
• Incremental twin support vector machines

zbMATH Keywords

classificationsupport vector machinestochastic gradient descentlarge-scale problemtwin support vector machine

Mathematics Subject Classification ID

Classification and discrimination; cluster analysis (statistical aspects) (62H30) Learning and adaptive systems in artificial intelligence (68T05)

Cites Work

• LIBLINEAR: a library for large linear classification
• TPMSVM: A novel twin parametric-margin support vector machine for pattern recognition
• SSVM: A smooth support vector machine for classification
• Pegasos: primal estimated sub-gradient solver for SVM
• MLTSVM: a novel twin support vector machine to multi-label learning
• Support-vector networks
• Title not available (Why is that?)
• Pattern classification.
• Title not available (Why is that?)
• Optimal kernel selection in twin support vector machines
• Twin support vector machine with Universum data
• Robust twin support vector machine for pattern classification
• Large-scale linear nonparallel support vector machine solver
• Almost sure convergence of a stochastic approximation process in a convex set
• A novel Frank-Wolfe algorithm. Analysis and applications to large-scale SVM training
• The new interpretation of support vector machines on statistical learning theory
• A GA-based model selection for smooth twin parametric-margin support vector machine
• Nonparallel hyperplane support vector machine for binary classification problems
• An efficient weighted Lagrangian twin support vector machine for imbalanced data classification
• A coordinate descent margin based-twin support vector machine for classification
• Stochastic gradient descent with Barzilai-Borwein update step for SVM

Cited In (10)

• Comprehensive review on twin support vector machines
• A novel twin parametric support vector machine for large scale problem
• General twin support vector machine with pinball loss function
• R-CTSVM+: robust capped \(\mathrm{L}_1\)-norm twin support vector machine with privileged information
• Pegasos: primal estimated sub-gradient solver for SVM
• Business analytics for corporate risk management and performance improvement
• Locally finite distance clustering with discriminative information
• Title not available (Why is that?)
• A novel regularization method for decorrelation learning of non-parallel hyperplanes
• Smooth augmented Lagrangian method for twin bounded support vector machine

Uses Software

• LIBSVM
• LIBLINEAR
• Pegasos
• SSVM
• TPMSVM
• RSVM
• MLTSVM

This page was built for publication: Insensitive stochastic gradient twin support vector machines for large scale problems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2198233)