Artificial data and language theory

Antoine Cornuéjols 1 Franck Thollard
1 TANC - Algorithmic number theory for cryptology
LIX - Laboratoire d'informatique de l'École polytechnique [Palaiseau], Inria Saclay - Ile de France, X - École polytechnique, CNRS - Centre National de la Recherche Scientifique : UMR7161
Abstract : Tests on artificial data sets are important in order to systematically assess the merits and limits of learning algorithms. This requires that distributions over both the hypothesis space and the instance space be controlled, which, in turn, implies the definition of control parameters. This paper surveys the control parameters that have been used in the context of grammatical inference, and provides directions for further identification of relevant parameters.
Type de document :
Communication dans un congrès
GI workshop 2006. Grammatical inference: workshop on open problems and new directions, Nov 2005, Saint-Etienne, France, 2005
Liste complète des métadonnées

https://hal.inria.fr/inria-00119756
Contributeur : Antoine Cornuéjols <>
Soumis le : lundi 11 décembre 2006 - 23:19:46
Dernière modification le : jeudi 10 mai 2018 - 02:06:31

Identifiants

  • HAL Id : inria-00119756, version 1

Collections

Citation

Antoine Cornuéjols, Franck Thollard. Artificial data and language theory. GI workshop 2006. Grammatical inference: workshop on open problems and new directions, Nov 2005, Saint-Etienne, France, 2005. 〈inria-00119756〉

Partager

Métriques

Consultations de la notice

137