Header logo is ei

Unifying Divergence Minimization and Statistical Inference Via Convex Duality

2006

Conference Paper

ei


In this paper we unify divergence minimization and statistical inference by means of convex duality. In the process of doing so, we prove that the dual of approximate maximum entropy estimation is maximum a posteriori estimation as a special case. Moreover, our treatment leads to stability and convergence bounds for many statistical learning problems. Finally, we show how an algorithm by Zhang can be used to solve this class of optimization problems efficiently.

Author(s): Altun, Y. and Smola, AJ.
Book Title: Learning Theory
Journal: Learning Theory: 19th Annual Conference on Learning Theory (COLT 2006)
Pages: 139-153
Year: 2006
Month: June
Day: 0
Editors: Lugosi, G. , H.-U. Simon
Publisher: Springer

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

DOI: 10.1007/11776420_13
Event Name: 19th Annual Conference on Learning Theory (COLT 2006)
Event Place: Pittsburgh, PA, USA

Address: Berlin, Germany
Digital: 0
Language: en
Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik

Links: Web

BibTex

@inproceedings{5704,
  title = {Unifying Divergence Minimization and Statistical Inference Via Convex Duality},
  author = {Altun, Y. and Smola, AJ.},
  journal = {Learning Theory: 19th Annual Conference on Learning Theory (COLT 2006)},
  booktitle = {Learning Theory},
  pages = {139-153},
  editors = {Lugosi, G. , H.-U. Simon},
  publisher = {Springer},
  organization = {Max-Planck-Gesellschaft},
  school = {Biologische Kybernetik},
  address = {Berlin, Germany},
  month = jun,
  year = {2006},
  month_numeric = {6}
}