Localized Multiple Kernel Learning—A Convex Approach

Abstract

We propose a localized approach to multiple kernel learning that can be formulated as a convex optimization problem over a given cluster structure. For which we obtain generalization error guarantees and derive an optimization algorithm based on the Fenchel dual representation. Experiments on real-world datasets from the application domains of computational biology and computer vision show that convex localized multiple kernel learning can achieve higher prediction accuracies than its global and non-convex local counterparts.

Cite

Text

Lei et al. "Localized Multiple Kernel Learning—A Convex Approach." Proceedings of The 8th Asian Conference on Machine Learning, 2016.

Markdown

[Lei et al. "Localized Multiple Kernel Learning—A Convex Approach." Proceedings of The 8th Asian Conference on Machine Learning, 2016.](https://mlanthology.org/acml/2016/lei2016acml-localized/)

BibTeX

@inproceedings{lei2016acml-localized,
  title     = {{Localized Multiple Kernel Learning—A Convex Approach}},
  author    = {Lei, Yunwen and Binder, Alexander and Dogan, Urun and Kloft, Marius},
  booktitle = {Proceedings of The 8th Asian Conference on Machine Learning},
  year      = {2016},
  pages     = {81-96},
  volume    = {63},
  url       = {https://mlanthology.org/acml/2016/lei2016acml-localized/}
}