Empirical Inference

Support Vector Machines as Probabilistic Models

2011

Conference Paper

ei


We show how the SVM can be viewed as a maximum likelihood estimate of a class of probabilistic models. This model class can be viewed as a reparametrization of the SVM in a similar vein to the v-SVM reparametrizing the classical (C-)SVM. It is not discriminative, but has a non-uniform marginal. We illustrate the benefits of this new view by rederiving and re-investigating two established SVM-related algorithms.

Author(s): Franc, V. and Zien, A. and Schölkopf, B.
Book Title: Proceedings of the 28th International Conference on Machine Learning
Pages: 665-672
Year: 2011
Month: July
Day: 0
Editors: L Getoor and T Scheffer
Publisher: International Machine Learning Society

Department(s): Empirical Inference
Bibtex Type: Conference Paper (inproceedings)

Event Name: ICML 2011
Event Place: Bellevue, WA, USA

Address: Madison, WI, USA
ISBN: 978-1-450-30619-5

Links: PDF
Web

BibTex

@inproceedings{FrancZS2011,
  title = {Support Vector Machines as Probabilistic Models},
  author = {Franc, V. and Zien, A. and Sch{\"o}lkopf, B.},
  booktitle = {Proceedings of the 28th International Conference on Machine Learning},
  pages = {665-672},
  editors = {L Getoor and T Scheffer},
  publisher = {International Machine Learning Society},
  address = {Madison, WI, USA},
  month = jul,
  year = {2011},
  doi = {},
  month_numeric = {7}
}