Header logo is

Trading Convexity for Scalability

2006

Conference Paper

ei


Convex learning algorithms, such as Support Vector Machines (SVMs), are often seen as highly desirable because they offer strong practical properties and are amenable to theoretical analysis. However, in this work we show how non-convexity can provide scalability advantages over convexity. We show how concave-convex programming can be applied to produce (i) faster SVMs where training errors are no longer support vectors, and (ii) much faster Transductive SVMs.

Author(s): Collobert, R. and Sinz, F. and Weston, J. and Bottou, L.
Book Title: ICML 2006
Journal: Proceedings of the 23rd International Conference on Machine Learning (ICML 2006)
Pages: 201-208
Year: 2006
Month: June
Day: 0
Editors: Cohen, W. W., A. Moore
Publisher: ACM Press

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

DOI: 10.1145/1143844.1143870
Event Name: 23rd International Conference on Machine Learning
Event Place: Pittsburgh, PA, USA

Address: New York, NY, USA
Digital: 0
Institution: Association for Computing Machinery
Language: en
Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik

Links: PDF
Web

BibTex

@inproceedings{3917,
  title = {Trading Convexity for Scalability},
  author = {Collobert, R. and Sinz, F. and Weston, J. and Bottou, L.},
  journal = {Proceedings of the 23rd International Conference on Machine Learning (ICML 2006)},
  booktitle = {ICML 2006},
  pages = {201-208},
  editors = {Cohen, W. W., A. Moore},
  publisher = {ACM Press},
  organization = {Max-Planck-Gesellschaft},
  institution = {Association for Computing Machinery},
  school = {Biologische Kybernetik},
  address = {New York, NY, USA},
  month = jun,
  year = {2006},
  doi = {10.1145/1143844.1143870},
  month_numeric = {6}
}