Multi-Class Classification with Maximum Margin Multiple Kernel

Corinna Cortes, Mehryar Mohri, Afshin Rostamizadeh
; Proceedings of the 30th International Conference on Machine Learning, PMLR 28(3):46-54, 2013.

Abstract

We present a new algorithm for multi-class classification with multiple kernels. Our algorithm is based on a natural notion of the multi-class margin of a kernel. We show that larger values of this quantity guarantee the existence of an accurate multi-class predictor and also define a family of multiple kernel algorithms based on the maximization of the multi-class margin of a kernel (M^3K). We present an extensive theoretical analysis in support of our algorithm, including novel multi-class Rademacher complexity margin bounds. Finally, we also report the results of a series of experiments with several data sets, including comparisons where we improve upon the performance of state-of-the-art algorithms both in binary and multi-class classification with multiple kernels.

Cite this Paper


BibTeX
@InProceedings{pmlr-v28-cortes13, title = {Multi-Class Classification with Maximum Margin Multiple Kernel}, author = {Corinna Cortes and Mehryar Mohri and Afshin Rostamizadeh}, booktitle = {Proceedings of the 30th International Conference on Machine Learning}, pages = {46--54}, year = {2013}, editor = {Sanjoy Dasgupta and David McAllester}, volume = {28}, number = {3}, series = {Proceedings of Machine Learning Research}, address = {Atlanta, Georgia, USA}, month = {17--19 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v28/cortes13.pdf}, url = {http://proceedings.mlr.press/v28/cortes13.html}, abstract = {We present a new algorithm for multi-class classification with multiple kernels. Our algorithm is based on a natural notion of the multi-class margin of a kernel. We show that larger values of this quantity guarantee the existence of an accurate multi-class predictor and also define a family of multiple kernel algorithms based on the maximization of the multi-class margin of a kernel (M^3K). We present an extensive theoretical analysis in support of our algorithm, including novel multi-class Rademacher complexity margin bounds. Finally, we also report the results of a series of experiments with several data sets, including comparisons where we improve upon the performance of state-of-the-art algorithms both in binary and multi-class classification with multiple kernels.} }
Endnote
%0 Conference Paper %T Multi-Class Classification with Maximum Margin Multiple Kernel %A Corinna Cortes %A Mehryar Mohri %A Afshin Rostamizadeh %B Proceedings of the 30th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2013 %E Sanjoy Dasgupta %E David McAllester %F pmlr-v28-cortes13 %I PMLR %J Proceedings of Machine Learning Research %P 46--54 %U http://proceedings.mlr.press %V 28 %N 3 %W PMLR %X We present a new algorithm for multi-class classification with multiple kernels. Our algorithm is based on a natural notion of the multi-class margin of a kernel. We show that larger values of this quantity guarantee the existence of an accurate multi-class predictor and also define a family of multiple kernel algorithms based on the maximization of the multi-class margin of a kernel (M^3K). We present an extensive theoretical analysis in support of our algorithm, including novel multi-class Rademacher complexity margin bounds. Finally, we also report the results of a series of experiments with several data sets, including comparisons where we improve upon the performance of state-of-the-art algorithms both in binary and multi-class classification with multiple kernels.
RIS
TY - CPAPER TI - Multi-Class Classification with Maximum Margin Multiple Kernel AU - Corinna Cortes AU - Mehryar Mohri AU - Afshin Rostamizadeh BT - Proceedings of the 30th International Conference on Machine Learning PY - 2013/02/13 DA - 2013/02/13 ED - Sanjoy Dasgupta ED - David McAllester ID - pmlr-v28-cortes13 PB - PMLR SP - 46 DP - PMLR EP - 54 L1 - http://proceedings.mlr.press/v28/cortes13.pdf UR - http://proceedings.mlr.press/v28/cortes13.html AB - We present a new algorithm for multi-class classification with multiple kernels. Our algorithm is based on a natural notion of the multi-class margin of a kernel. We show that larger values of this quantity guarantee the existence of an accurate multi-class predictor and also define a family of multiple kernel algorithms based on the maximization of the multi-class margin of a kernel (M^3K). We present an extensive theoretical analysis in support of our algorithm, including novel multi-class Rademacher complexity margin bounds. Finally, we also report the results of a series of experiments with several data sets, including comparisons where we improve upon the performance of state-of-the-art algorithms both in binary and multi-class classification with multiple kernels. ER -
APA
Cortes, C., Mohri, M. & Rostamizadeh, A.. (2013). Multi-Class Classification with Maximum Margin Multiple Kernel. Proceedings of the 30th International Conference on Machine Learning, in PMLR 28(3):46-54

Related Material