ML Anthology

Gaps in Support Vector Optimization

Nikolas List, Don R. Hush, Clint Scovel, Ingo Steinwart
COLT 2007 pp. 336-348
doi:10.1007/978-3-540-72927-3_25 /colt/2007/list2007colt-gaps/

Abstract

We show that the stopping criteria used in many support vector machine (SVM) algorithms working on the dual can be interpreted as primal optimality bounds which in turn are known to be important for the statistical analysis of SVMs. To this end we revisit the duality theory underlying the derivation of the dual and show that in many interesting cases primal optimality bounds are the same as known dual optimality bounds.

PDF COLT Semantic Scholar

Cite

Text

List et al. "Gaps in Support Vector Optimization." Annual Conference on Computational Learning Theory, 2007. doi:10.1007/978-3-540-72927-3_25

Markdown

[List et al. "Gaps in Support Vector Optimization." Annual Conference on Computational Learning Theory, 2007.](https://mlanthology.org/colt/2007/list2007colt-gaps/) doi:10.1007/978-3-540-72927-3_25

BibTeX

@inproceedings{list2007colt-gaps,
  title     = {{Gaps in Support Vector Optimization}},
  author    = {List, Nikolas and Hush, Don R. and Scovel, Clint and Steinwart, Ingo},
  booktitle = {Annual Conference on Computational Learning Theory},
  year      = {2007},
  pages     = {336-348},
  doi       = {10.1007/978-3-540-72927-3_25},
  url       = {https://mlanthology.org/colt/2007/list2007colt-gaps/}
}