On-chip learning of FPGA-inspired neural nets - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2001

On-chip learning of FPGA-inspired neural nets

Bernard Girau

Résumé

Neural networks are usually considered as naturally parallel computing models. But the number of operators and the complex connection graphs of standard neural models can not be handled by digital hardware devices. A new theoretical and practical framework allows to reconcile simple hardware topologies with complex neural architectures: Field Programmable Neural Arrays (FPNA) lead to powerful neural architectures that are easy to map onto digital hardware, thanks to a simplified topology and an original data exchange scheme. This paper focuses on a class of synchronous FPNAs, for which an efficient implementation with on-chip learning is described. Application and implementation results are rapidly discussed.

Domaines

Autre [cs.OH]
Fichier non déposé

Dates et versions

inria-00100467 , version 1 (26-09-2006)

Identifiants

  • HAL Id : inria-00100467 , version 1

Citer

Bernard Girau. On-chip learning of FPGA-inspired neural nets. International Joint Conference on Neural Networks - IJCNN'01, Jul 2001, Washington, USA, 6 p. ⟨inria-00100467⟩
134 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More