Max-Margin Ratio Machine

Suicheng Gu, Yuhong Guo
Proceedings of the Asian Conference on Machine Learning, PMLR 25:145-157, 2012.

Abstract

In this paper, we investigate the problem of exploiting global information to improve the performance of SVMs on large scale classification problems. We first present a unified general framework for the existing min-max machine methods in terms of within-class dispersions and between-class dispersions. By defining a new within-class dispersion measure, we then propose a novel max-margin ratio machine (MMRM) method that can be formulated as a linear programming problem with scalability for large data sets. Kernels can be easily incorporated into our method to address non-linear classification problems. Our empirical results show that the proposed MMRM approach achieves promising results on large data sets.

Cite this Paper


BibTeX
@InProceedings{pmlr-v25-gu12, title = {Max-Margin Ratio Machine}, author = {Gu, Suicheng and Guo, Yuhong}, booktitle = {Proceedings of the Asian Conference on Machine Learning}, pages = {145--157}, year = {2012}, editor = {Hoi, Steven C. H. and Buntine, Wray}, volume = {25}, series = {Proceedings of Machine Learning Research}, address = {Singapore Management University, Singapore}, month = {04--06 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v25/gu12/gu12.pdf}, url = {https://proceedings.mlr.press/v25/gu12.html}, abstract = {In this paper, we investigate the problem of exploiting global information to improve the performance of SVMs on large scale classification problems. We first present a unified general framework for the existing min-max machine methods in terms of within-class dispersions and between-class dispersions. By defining a new within-class dispersion measure, we then propose a novel max-margin ratio machine (MMRM) method that can be formulated as a linear programming problem with scalability for large data sets. Kernels can be easily incorporated into our method to address non-linear classification problems. Our empirical results show that the proposed MMRM approach achieves promising results on large data sets.} }
Endnote
%0 Conference Paper %T Max-Margin Ratio Machine %A Suicheng Gu %A Yuhong Guo %B Proceedings of the Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2012 %E Steven C. H. Hoi %E Wray Buntine %F pmlr-v25-gu12 %I PMLR %P 145--157 %U https://proceedings.mlr.press/v25/gu12.html %V 25 %X In this paper, we investigate the problem of exploiting global information to improve the performance of SVMs on large scale classification problems. We first present a unified general framework for the existing min-max machine methods in terms of within-class dispersions and between-class dispersions. By defining a new within-class dispersion measure, we then propose a novel max-margin ratio machine (MMRM) method that can be formulated as a linear programming problem with scalability for large data sets. Kernels can be easily incorporated into our method to address non-linear classification problems. Our empirical results show that the proposed MMRM approach achieves promising results on large data sets.
RIS
TY - CPAPER TI - Max-Margin Ratio Machine AU - Suicheng Gu AU - Yuhong Guo BT - Proceedings of the Asian Conference on Machine Learning DA - 2012/11/17 ED - Steven C. H. Hoi ED - Wray Buntine ID - pmlr-v25-gu12 PB - PMLR DP - Proceedings of Machine Learning Research VL - 25 SP - 145 EP - 157 L1 - http://proceedings.mlr.press/v25/gu12/gu12.pdf UR - https://proceedings.mlr.press/v25/gu12.html AB - In this paper, we investigate the problem of exploiting global information to improve the performance of SVMs on large scale classification problems. We first present a unified general framework for the existing min-max machine methods in terms of within-class dispersions and between-class dispersions. By defining a new within-class dispersion measure, we then propose a novel max-margin ratio machine (MMRM) method that can be formulated as a linear programming problem with scalability for large data sets. Kernels can be easily incorporated into our method to address non-linear classification problems. Our empirical results show that the proposed MMRM approach achieves promising results on large data sets. ER -
APA
Gu, S. & Guo, Y.. (2012). Max-Margin Ratio Machine. Proceedings of the Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 25:145-157 Available from https://proceedings.mlr.press/v25/gu12.html.

Related Material