Header logo is

A Kernel Approach to Comparing Distributions

2007

Conference Paper

ei


We describe a technique for comparing distributions without the need for density estimation as an intermediate step. Our approach relies on mapping the distributions into a Reproducing Kernel Hilbert Space. We apply this technique to construct a two-sample test, which is used for determining whether two sets of observations arise from the same distribution. We use this test in attribute matching for databases using the Hungarian marriage method, where it performs strongly. We also demonstrate excellent performance when comparing distributions over graphs, for which no alternative tests currently exist.

Author(s): Gretton, A. and Borgwardt, KM. and Rasch, M. and Schölkopf, B. and Smola, AJ.
Book Title: Proceedings of the 22. AAAI Conference on Artificial Intelligence
Journal: Proceedings of the Twenty-Second AAAI Conference on Artificial Intelligence (AAAI-07)
Pages: 1637-1641
Year: 2007
Month: July
Day: 0
Publisher: AAAI Press

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

Event Name: Twenty-Second AAAI Conference on Artificial Intelligence (AAAI 2007)
Event Place: Vancouver, BC, Canada

Address: Menlo Park, CA, USA
Digital: 0
Institution: Association for the Advancement of Artificial Intelligence
ISBN: 978-1-577-35323-2
Language: en
Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik

Links: PDF
Web

BibTex

@inproceedings{4426,
  title = {A Kernel Approach to Comparing Distributions},
  author = {Gretton, A. and Borgwardt, KM. and Rasch, M. and Sch{\"o}lkopf, B. and Smola, AJ.},
  journal = {Proceedings of the Twenty-Second AAAI Conference on Artificial Intelligence (AAAI-07)},
  booktitle = {Proceedings of the 22. AAAI Conference on Artificial Intelligence},
  pages = {1637-1641},
  publisher = {AAAI Press},
  organization = {Max-Planck-Gesellschaft},
  institution = {Association for the Advancement of Artificial Intelligence},
  school = {Biologische Kybernetik},
  address = {Menlo Park, CA, USA},
  month = jul,
  year = {2007},
  doi = {},
  month_numeric = {7}
}