Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

TV-SVM: Support Vector Machine with Total Variational Regularization

Abstract: 

To leverage the spatial relationship of lattice data, such as images, we introduce total variational (TV) regularization into support vector machines (SVM), called TV-SVM. TV-SVM encourages local smoothness and sparsity in gradient domain of the learned parameters. TV-SVM is optimized via the alternating direction method of multipliers (ADMM) algorithm and is significantly better than (Linear) SVM for image classifications.

up
1 user has voted: Zhendong Zhang

Paper Details

Authors:
Zhendong Zhang,Cheolkon Jung
Submitted On:
13 April 2018 - 12:30pm
Short Link:
Type:
Poster
Event:
Presenter's Name:
Zhendong Zhang
Paper Code:
1082
Document Year:
2018
Cite

Document Files

ICASSP2018poster_TVSVM_final.pdf

(22 downloads)

Subscribe

[1] Zhendong Zhang,Cheolkon Jung, "TV-SVM: Support Vector Machine with Total Variational Regularization", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/2728. Accessed: May. 25, 2018.
@article{2728-18,
url = {http://sigport.org/2728},
author = {Zhendong Zhang;Cheolkon Jung },
publisher = {IEEE SigPort},
title = {TV-SVM: Support Vector Machine with Total Variational Regularization},
year = {2018} }
TY - EJOUR
T1 - TV-SVM: Support Vector Machine with Total Variational Regularization
AU - Zhendong Zhang;Cheolkon Jung
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/2728
ER -
Zhendong Zhang,Cheolkon Jung. (2018). TV-SVM: Support Vector Machine with Total Variational Regularization. IEEE SigPort. http://sigport.org/2728
Zhendong Zhang,Cheolkon Jung, 2018. TV-SVM: Support Vector Machine with Total Variational Regularization. Available at: http://sigport.org/2728.
Zhendong Zhang,Cheolkon Jung. (2018). "TV-SVM: Support Vector Machine with Total Variational Regularization." Web.
1. Zhendong Zhang,Cheolkon Jung. TV-SVM: Support Vector Machine with Total Variational Regularization [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/2728