Tools & Methodology for Hybrid Neuro-Symbolic Models

Frédéric Alexandre 1
1 CORTEX - Neuromimetic intelligence
INRIA Lorraine, LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
Abstract : Hybrid Neuro-Symbolic models are nowadays commonly used for intelligent information processing. Nevertheless, no serious theoretical study was performed in the past to settle a solid basis to hybridization. From this idea, a european ESPRIT project, the project MIX, funded by the European Union has been proposed, in order to study hybridization, develop tools and propose strategies to better analyse and emulate hybrid models. We explore in this paper the main results (theoretical and practical) of this project. They correspond to a toolbox for the implementation and communication of neural and symbolic agents and to some taxinomic aspects for the better classification and understanding of hybrid models. Thanks to these results, we were able to propose a general methodology for hybridization and to assess it thanks to real-world applications. Both aspects are also evoked in the paper.
Type de document :
Communication dans un congrès
IEEE Systems, Man & Cybernetics. IMACS Int. Conf. on Computational Engineering in Systems Applications - CESA'98, 1998, Nabeul (Tunisie), 1998
Liste complète des métadonnées

https://hal.inria.fr/inria-00108034
Contributeur : Publications Loria <>
Soumis le : jeudi 19 octobre 2006 - 15:40:17
Dernière modification le : jeudi 11 janvier 2018 - 06:19:48

Identifiants

  • HAL Id : inria-00108034, version 1

Collections

Citation

Frédéric Alexandre. Tools & Methodology for Hybrid Neuro-Symbolic Models. IEEE Systems, Man & Cybernetics. IMACS Int. Conf. on Computational Engineering in Systems Applications - CESA'98, 1998, Nabeul (Tunisie), 1998. 〈inria-00108034〉

Partager

Métriques

Consultations de la notice

139