ML Anthology

Trading Convexity for Scalability

Ronan Collobert, Fabian H. Sinz, Jason Weston, Léon Bottou
ICML 2006 pp. 201-208
doi:10.1145/1143844.1143870 /icml/2006/collobert2006icml-trading/

Abstract

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.

PDF ICML Semantic Scholar

Cite

Text

Collobert et al. "Trading Convexity for Scalability." International Conference on Machine Learning, 2006. doi:10.1145/1143844.1143870

Markdown

[Collobert et al. "Trading Convexity for Scalability." International Conference on Machine Learning, 2006.](https://mlanthology.org/icml/2006/collobert2006icml-trading/) doi:10.1145/1143844.1143870

BibTeX

@inproceedings{collobert2006icml-trading,
  title     = {{Trading Convexity for Scalability}},
  author    = {Collobert, Ronan and Sinz, Fabian H. and Weston, Jason and Bottou, Léon},
  booktitle = {International Conference on Machine Learning},
  year      = {2006},
  pages     = {201-208},
  doi       = {10.1145/1143844.1143870},
  url       = {https://mlanthology.org/icml/2006/collobert2006icml-trading/}
}