Skip to Main content Skip to Navigation
Conference papers

Bayesian networks : a better than frequentist approach for parametrization, and a more accurate structural complexity measure than the number of parameters

Sylvain Gelly 1 Olivier Teytaud 1
1 TANC - Algorithmic number theory for cryptology
Inria Saclay - Ile de France, LIX - Laboratoire d'informatique de l'École polytechnique [Palaiseau]
Abstract : We propose and justify a better-than-frequentist approach for bayesian network parametrization, and propose a structural entropy term that more precisely quantifies the complexity of a BN than the number of parameters. Algorithms for BN learning are deduced.
Document type :
Conference papers
Complete list of metadata

https://hal.inria.fr/inria-00000541
Contributor : Olivier Teytaud <>
Submitted on : Monday, October 31, 2005 - 10:34:05 PM
Last modification on : Wednesday, March 27, 2019 - 4:41:29 PM
Long-term archiving on: : Friday, April 2, 2010 - 6:14:18 PM

File

Identifiers

  • HAL Id : inria-00000541, version 1

Collections

Citation

Sylvain Gelly, Olivier Teytaud. Bayesian networks : a better than frequentist approach for parametrization, and a more accurate structural complexity measure than the number of parameters. CAP, 2005, Nice, 16 p. ⟨inria-00000541⟩

Share

Metrics

Record views

227

Files downloads

218