Multiple kernel learning with data augmentation

Khanh Nguyen, Trung Le, Vu Nguyen, Tu Dinh Nguyen , Dinh Phung

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearch

4 Citations (Scopus)

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-aposteriori (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.
Original languageEnglish
Title of host publication8th Asian Conference on Machine Learning (ACML)
Subtitle of host publication16-18 November 2016, The University of Waikato, Hamilton, New Zealand
EditorsBob Durrant, Kee-Eung Kim
Place of PublicationSheffield UK
PublisherProceedings of Machine Learning Research (PMLR)
Pages49-64
Number of pages16
Volume63
ISBN (Electronic)2640-3498
Publication statusPublished - 2016
Externally publishedYes
EventAsian Conference on Machine Learning 2016 - Waikato, New Zealand
Duration: 16 Nov 201618 Nov 2016
Conference number: 8th
http://www.acml-conf.org/2016/
http://proceedings.mlr.press/v63/ (Proceedings)

Publication series

NameJournal of Machine Learning Research
PublisherJournal of Machine Learning Research (JMLR)
ISSN (Print)1532-4435

Conference

ConferenceAsian Conference on Machine Learning 2016
Abbreviated titleACML 2016
CountryNew Zealand
CityWaikato
Period16/11/1618/11/16
Internet address

Keywords

  • Data Augmentation
  • Multiple Kernel Learning
  • Stochastic Gradient Descent

Cite this

Nguyen, K., Le, T., Nguyen, V., Nguyen , T. D., & Phung, D. (2016). Multiple kernel learning with data augmentation. In B. Durrant, & K-E. Kim (Eds.), 8th Asian Conference on Machine Learning (ACML): 16-18 November 2016, The University of Waikato, Hamilton, New Zealand (Vol. 63, pp. 49-64). (Journal of Machine Learning Research). Proceedings of Machine Learning Research (PMLR). http://proceedings.mlr.press/v63/nguyen19.pdf