Plasticity in memristive devices for Spiking Neural Networks

Frontiers in Neuroscience

卷:9

Olivier eBichler¹  Christian eGamrat¹  Julie eGrollier²  Dominique eVuillaume²  Selina eLa Barbera²  Sören eBoyn²  Fabien eAlibart²  Christian G Mayr³  Heidemarie eSchmidt⁴  Adrien eVincent⁵  Damien eQuerlioz⁵  Teresa eSerrano-Gotarredona⁶  Bernabe eLinares-Barranco⁶  Gwendal eLecerf⁷  Jean eTomas⁷  Sylvain eSaïghi⁷ 

[1] CEA LIST;

[2] CNRS;

[3] ETH Zurich;

[4] Technische Universität Chemnitz;

[5] Univ. Paris-Sud;

[6] Universidad de Sevilla;

[7] Université de Bordeaux;

关键词: plasticity;    neuromorphic engineering;    Memristor;    Memristive device;    Hardware neural network;   

DOI  :  10.3389/fnins.2015.00051

来源: DOAJ

【 摘 要 】

Memristive devices present a new device technology allowing for the realization of compact nonvolatile memories. Some of them are already in the process of industrialization. Additionally, they exhibit complex multilevel and plastic behaviors, which make them good candidates for the implementation of artificial synapses in neuromorphic engineering. However, memristive effects rely on diverse physical mechanisms, and their plastic behaviors differ strongly from one technology to another. Here, we present measurements performed on different memristive devices and the opportunities that they provide. We show that they can be used to implement different learning rules whose properties emerge directly from device physics: real time or accelerated operation, deterministic or stochastic behavior, long term or short term plasticity. We then discuss how such devices might be integrated into a complete architecture. These results highlight that there is no unique way to exploit memristive devices in neuromorphic systems. Understanding and embracing device physics is the key for their optimal use.

【 授权许可】

Unknown