In-Memory and Error-Immune Differential RRAM Implementation of Binarized Deep Neural Networks - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2018

In-Memory and Error-Immune Differential RRAM Implementation of Binarized Deep Neural Networks

Résumé

RRAM-based in-Memory Computing is an exciting road for implementing highly energy efficient neural networks. This vision is however challenged by RRAM variability, as the efficient implementation of in-memory computing does not allow error correction. In this work, we fabricated and tested a differential HfO 2-based memory structure and its associated sense circuitry, which are ideal for in-memory computing. For the first time, we show that our approach achieves the same reliability benefits as error correction, but without any CMOS overhead. We show, also for the first time, that it can naturally implement Binarized Deep Neural Networks, a very recent development of Artificial Intelligence, with extreme energy efficiency, and that the system is fully satisfactory for image recognition applications. Finally, we evidence how the extra reliability provided by the differential memory allows programming the devices in low voltage conditions, where they feature high endurance of billions of cycles.
Fichier principal
Vignette du fichier
Bocquet_IEDM2018.pdf (2.87 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02011124 , version 1 (07-02-2019)

Identifiants

Citer

Marc Bocquet, T. Hirztlin, J.-O. Klein, E. Nowak, E. Vianello, et al.. In-Memory and Error-Immune Differential RRAM Implementation of Binarized Deep Neural Networks. 2018 IEEE International Electron Devices Meeting (IEDM), Dec 2018, San Francisco, United States. pp.20.6.1-20.6.4, ⟨10.1109/IEDM.2018.8614639⟩. ⟨hal-02011124⟩
218 Consultations
99 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More