inria-00000541, version 1
Bayesian networks : a better than frequentist approach for parametrization, and a more accurate structural complexity measure than the number of parameters
Sylvain Gelly 1Olivier Teytaud
1
CAP (2005) 16 pages
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.
- 1: TAO (INRIA Futurs)
- INRIA – CNRS : UMR8623 – Université Paris XI - Paris Sud
- Domain : Computer Science/Learning
- inria-00000541, version 1
- http://hal.inria.fr/inria-00000541
- oai:hal.inria.fr:inria-00000541
- From: Olivier Teytaud
- Submitted on: Monday, 31 October 2005 22:34:05
- Updated on: Tuesday, 13 December 2005 10:46:14
Associated documents
Export