Analog implementation of encoded neural networks - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2013

Analog implementation of encoded neural networks

Résumé

Encoded neural networks mix the principles of associative memories and error-correcting decoders. Their storage capacity has been shown to be much larger than Hopfield Neural Networks’. This paper introduces an analog implementation of this new type of network. The proposed circuit has been designed for the 1V supply ST CMOS 65nm process. It consumes 1165 times less energy than a digital equivalent circuit while being 2.7 times more efficient in terms of combined speed and surface.
Fichier non déposé

Dates et versions

hal-00836649 , version 1 (21-06-2013)

Identifiants

  • HAL Id : hal-00836649 , version 1

Citer

Benoit Larras, Cyril Lahuec, Matthieu Arzel, Fabrice Seguin. Analog implementation of encoded neural networks. ISCAS 2013 : IEEE International Symposium on Circuits and Systems, May 2013, Beijing, China. pp.1 - 4. ⟨hal-00836649⟩
234 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More