Multiple Kernel Learning with Data Augmentation

Khanh Nguyen, Trung Le, Vu Nguyen, Tu Nguyen, Dinh Phung
Proceedings of The 8th Asian Conference on Machine Learning, PMLR 63:49-64, 2016.

Abstract

The motivations of multiple kernel learning (MKL) approach are to increase kernel expressiveness capacity and to avoid the expensive grid search over a wide spectrum of kernels. A large amount of work has been proposed to improve the MKL in terms of the computational cost and the sparsity of the solution. However, these studies still either require an expensive grid search on the model parameters or scale unsatisfactorily with the numbers of kernels and training samples. In this paper, we address these issues by conjoining MKL, Stochastic Gradient Descent (SGD) framework, and data augmentation technique. The pathway of our proposed method is developed as follows. We first develop a maximum-a-posteriori (MAP) view for MKL under a probabilistic setting and described in a graphical model. This view allows us to develop data augmentation technique to make the inference for finding the optimal parameters feasible, as opposed to traditional approach of training MKL via convex optimization techniques. As a result, we can use the standard SGD framework to learn weight matrix and extend the model to support online learning. We validate our method on several benchmark datasets in both batch and online settings. The experimental results show that our proposed method can learn the parameters in a principled way to eliminate the expensive grid search while gaining a significant computational speedup comparing with the state-of-the-art baselines.

Cite this Paper

BibTeX
@InProceedings{pmlr-v63-nguyen19, title = {Multiple Kernel Learning with Data Augmentation}, author = {Nguyen, Khanh and Le, Trung and Nguyen, Vu and Nguyen, Tu and Phung, Dinh}, booktitle = {Proceedings of The 8th Asian Conference on Machine Learning}, pages = {49--64}, year = {2016}, editor = {Durrant, Robert J. and Kim, Kee-Eung}, volume = {63}, series = {Proceedings of Machine Learning Research}, address = {The University of Waikato, Hamilton, New Zealand}, month = {16--18 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v63/nguyen19.pdf}, url = {https://proceedings.mlr.press/v63/nguyen19.html}, abstract = {The motivations of multiple kernel learning (MKL) approach are to increase kernel expressiveness capacity and to avoid the expensive grid search over a wide spectrum of kernels. A large amount of work has been proposed to improve the MKL in terms of the computational cost and the sparsity of the solution. However, these studies still either require an expensive grid search on the model parameters or scale unsatisfactorily with the numbers of kernels and training samples. In this paper, we address these issues by conjoining MKL, Stochastic Gradient Descent (SGD) framework, and data augmentation technique. The pathway of our proposed method is developed as follows. We first develop a maximum-a-posteriori (MAP) view for MKL under a probabilistic setting and described in a graphical model. This view allows us to develop data augmentation technique to make the inference for finding the optimal parameters feasible, as opposed to traditional approach of training MKL via convex optimization techniques. As a result, we can use the standard SGD framework to learn weight matrix and extend the model to support online learning. We validate our method on several benchmark datasets in both batch and online settings. The experimental results show that our proposed method can learn the parameters in a principled way to eliminate the expensive grid search while gaining a significant computational speedup comparing with the state-of-the-art baselines.} }
Endnote
%0 Conference Paper %T Multiple Kernel Learning with Data Augmentation %A Khanh Nguyen %A Trung Le %A Vu Nguyen %A Tu Nguyen %A Dinh Phung %B Proceedings of The 8th Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2016 %E Robert J. Durrant %E Kee-Eung Kim %F pmlr-v63-nguyen19 %I PMLR %P 49--64 %U https://proceedings.mlr.press/v63/nguyen19.html %V 63 %X The motivations of multiple kernel learning (MKL) approach are to increase kernel expressiveness capacity and to avoid the expensive grid search over a wide spectrum of kernels. A large amount of work has been proposed to improve the MKL in terms of the computational cost and the sparsity of the solution. However, these studies still either require an expensive grid search on the model parameters or scale unsatisfactorily with the numbers of kernels and training samples. In this paper, we address these issues by conjoining MKL, Stochastic Gradient Descent (SGD) framework, and data augmentation technique. The pathway of our proposed method is developed as follows. We first develop a maximum-a-posteriori (MAP) view for MKL under a probabilistic setting and described in a graphical model. This view allows us to develop data augmentation technique to make the inference for finding the optimal parameters feasible, as opposed to traditional approach of training MKL via convex optimization techniques. As a result, we can use the standard SGD framework to learn weight matrix and extend the model to support online learning. We validate our method on several benchmark datasets in both batch and online settings. The experimental results show that our proposed method can learn the parameters in a principled way to eliminate the expensive grid search while gaining a significant computational speedup comparing with the state-of-the-art baselines.
RIS
TY - CPAPER TI - Multiple Kernel Learning with Data Augmentation AU - Khanh Nguyen AU - Trung Le AU - Vu Nguyen AU - Tu Nguyen AU - Dinh Phung BT - Proceedings of The 8th Asian Conference on Machine Learning DA - 2016/11/20 ED - Robert J. Durrant ED - Kee-Eung Kim ID - pmlr-v63-nguyen19 PB - PMLR DP - Proceedings of Machine Learning Research VL - 63 SP - 49 EP - 64 L1 - http://proceedings.mlr.press/v63/nguyen19.pdf UR - https://proceedings.mlr.press/v63/nguyen19.html AB - The motivations of multiple kernel learning (MKL) approach are to increase kernel expressiveness capacity and to avoid the expensive grid search over a wide spectrum of kernels. A large amount of work has been proposed to improve the MKL in terms of the computational cost and the sparsity of the solution. However, these studies still either require an expensive grid search on the model parameters or scale unsatisfactorily with the numbers of kernels and training samples. In this paper, we address these issues by conjoining MKL, Stochastic Gradient Descent (SGD) framework, and data augmentation technique. The pathway of our proposed method is developed as follows. We first develop a maximum-a-posteriori (MAP) view for MKL under a probabilistic setting and described in a graphical model. This view allows us to develop data augmentation technique to make the inference for finding the optimal parameters feasible, as opposed to traditional approach of training MKL via convex optimization techniques. As a result, we can use the standard SGD framework to learn weight matrix and extend the model to support online learning. We validate our method on several benchmark datasets in both batch and online settings. The experimental results show that our proposed method can learn the parameters in a principled way to eliminate the expensive grid search while gaining a significant computational speedup comparing with the state-of-the-art baselines. ER -
APA
Nguyen, K., Le, T., Nguyen, V., Nguyen, T. & Phung, D.. (2016). Multiple Kernel Learning with Data Augmentation. Proceedings of The 8th Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 63:49-64 Available from https://proceedings.mlr.press/v63/nguyen19.html.

Related Material