Online Learning of Structured Predictors with Multiple Kernels

Andre Filipe Torres Martins, Noah Smith, Eric Xing, Pedro Aguiar, Mario Figueiredo
Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, PMLR 15:507-515, 2011.

Abstract

Training structured predictors often requires a considerable time selecting features or tweaking the kernel. Multiple kernel learning (MKL) sidesteps this issue by embedding the kernel learning into the training procedure. Despite the recent progress towards efficiency of MKL algorithms, the structured output case remains an open research front. We propose a family of online algorithms able to tackle variants of MKL and group-LASSO, for which we show regret, convergence, and generalization bounds. Experiments on handwriting recognition and dependency parsing attest the success of the approach.

Cite this Paper


BibTeX
@InProceedings{pmlr-v15-martins11a, title = {Online Learning of Structured Predictors with Multiple Kernels}, author = {Martins, Andre Filipe Torres and Smith, Noah and Xing, Eric and Aguiar, Pedro and Figueiredo, Mario}, booktitle = {Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics}, pages = {507--515}, year = {2011}, editor = {Gordon, Geoffrey and Dunson, David and Dudík, Miroslav}, volume = {15}, series = {Proceedings of Machine Learning Research}, address = {Fort Lauderdale, FL, USA}, month = {11--13 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v15/martins11a/martins11a.pdf}, url = {https://proceedings.mlr.press/v15/martins11a.html}, abstract = {Training structured predictors often requires a considerable time selecting features or tweaking the kernel. Multiple kernel learning (MKL) sidesteps this issue by embedding the kernel learning into the training procedure. Despite the recent progress towards efficiency of MKL algorithms, the structured output case remains an open research front. We propose a family of online algorithms able to tackle variants of MKL and group-LASSO, for which we show regret, convergence, and generalization bounds. Experiments on handwriting recognition and dependency parsing attest the success of the approach.} }
Endnote
%0 Conference Paper %T Online Learning of Structured Predictors with Multiple Kernels %A Andre Filipe Torres Martins %A Noah Smith %A Eric Xing %A Pedro Aguiar %A Mario Figueiredo %B Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2011 %E Geoffrey Gordon %E David Dunson %E Miroslav Dudík %F pmlr-v15-martins11a %I PMLR %P 507--515 %U https://proceedings.mlr.press/v15/martins11a.html %V 15 %X Training structured predictors often requires a considerable time selecting features or tweaking the kernel. Multiple kernel learning (MKL) sidesteps this issue by embedding the kernel learning into the training procedure. Despite the recent progress towards efficiency of MKL algorithms, the structured output case remains an open research front. We propose a family of online algorithms able to tackle variants of MKL and group-LASSO, for which we show regret, convergence, and generalization bounds. Experiments on handwriting recognition and dependency parsing attest the success of the approach.
RIS
TY - CPAPER TI - Online Learning of Structured Predictors with Multiple Kernels AU - Andre Filipe Torres Martins AU - Noah Smith AU - Eric Xing AU - Pedro Aguiar AU - Mario Figueiredo BT - Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics DA - 2011/06/14 ED - Geoffrey Gordon ED - David Dunson ED - Miroslav Dudík ID - pmlr-v15-martins11a PB - PMLR DP - Proceedings of Machine Learning Research VL - 15 SP - 507 EP - 515 L1 - http://proceedings.mlr.press/v15/martins11a/martins11a.pdf UR - https://proceedings.mlr.press/v15/martins11a.html AB - Training structured predictors often requires a considerable time selecting features or tweaking the kernel. Multiple kernel learning (MKL) sidesteps this issue by embedding the kernel learning into the training procedure. Despite the recent progress towards efficiency of MKL algorithms, the structured output case remains an open research front. We propose a family of online algorithms able to tackle variants of MKL and group-LASSO, for which we show regret, convergence, and generalization bounds. Experiments on handwriting recognition and dependency parsing attest the success of the approach. ER -
APA
Martins, A.F.T., Smith, N., Xing, E., Aguiar, P. & Figueiredo, M.. (2011). Online Learning of Structured Predictors with Multiple Kernels. Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 15:507-515 Available from https://proceedings.mlr.press/v15/martins11a.html.

Related Material