ML Anthology

On the Dual Formulation of Regularized Linear Systems with Convex Risks

Tong Zhang
MLJ 2002 pp. 91-129
doi:10.1023/A:1012498226479 /mlj/2002/zhang2002mlj-dual/

Abstract

In this paper, we study a general formulation of linear prediction algorithms including a number of known methods as special cases. We describe a convex duality for this class of methods and propose numerical algorithms to solve the derived dual learning problem. We show that the dual formulation is closely related to online learning algorithms. Furthermore, by using this duality, we show that new learning methods can be obtained. Numerical examples will be given to illustrate various aspects of the newly proposed algorithms.

PDF MLJ Semantic Scholar

Cite

Text

Zhang. "On the Dual Formulation of Regularized Linear Systems with Convex Risks." Machine Learning, 2002. doi:10.1023/A:1012498226479

Markdown

[Zhang. "On the Dual Formulation of Regularized Linear Systems with Convex Risks." Machine Learning, 2002.](https://mlanthology.org/mlj/2002/zhang2002mlj-dual/) doi:10.1023/A:1012498226479

BibTeX

@article{zhang2002mlj-dual,
  title     = {{On the Dual Formulation of Regularized Linear Systems with Convex Risks}},
  author    = {Zhang, Tong},
  journal   = {Machine Learning},
  year      = {2002},
  pages     = {91-129},
  doi       = {10.1023/A:1012498226479},
  volume    = {46},
  url       = {https://mlanthology.org/mlj/2002/zhang2002mlj-dual/}
}