1. Memristive based device arrays combined with Spike based coding can enable efficient implementations of embedded neuromorphic circuits
- Author
-
Olivier Bichler, David Roclin, Christian Gamrat, Département d'Architectures, Conception et Logiciels Embarqués-LIST (DACLE-LIST), Laboratoire d'Intégration des Systèmes et des Technologies (LIST), Direction de Recherche Technologique (CEA) (DRT (CEA)), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Direction de Recherche Technologique (CEA) (DRT (CEA)), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA), The authors wishes to thank Barbara de Salvo, CEA LETI, for fruitful discussions.The work reported in this paper was partly supported by funding under the European Union Seventh Framework Program (FP7) undergrant agreement no 318597 (SYMONE), under the FrenchNational Research Agency (ANR) under grant agreement ANR12-BS03-010 (SYNAPTOR) and under grant agreement ANR–11–IDEX–0003-02 (NANODESIGN)., ANR-12-BS03-0010,SYNAPTOR,Transistor-synapse et circuits pour les architectures neuro-inspirées(2012), European Project: 318597,EC:FP7:ICT,FP7-ICT-2011-8,SYMONE(2012), and Laboratoire d'Intégration des Systèmes et des Technologies (LIST (CEA))
- Subjects
Computer science ,Biological neuron ,Neuromorphic Architectures ,law.invention ,Electron devices ,Embedded applications ,[SPI]Engineering Sciences [physics] ,law ,Electronic engineering ,[INFO]Computer Science [cs] ,Energy problem ,Implementation ,Neuromorphic circuits ,Electronic circuit ,Transistor ,Codes (symbols) ,CMOS integrated circuits ,Efficient implementation ,Reconfigurable hardware ,Energy efficiency ,Neuromorphic engineering ,Computer architecture ,Computing machines ,Device arrays ,Embedded application ,Coding (social sciences) ,Efficient energy use - Abstract
Conference of 61st IEEE International Electron Devices Meeting, IEDM 2015 ; Conference Date: 7 December 2015 Through 9 December 2015; Conference Code:119534; International audience; Since the rapid development of post-CMOS technologies in the last decade, there has been a growing interest in utilizing them for implementing neuromorphic or brain-like computing machines. Besides attempts to build realistic circuits that would mimic the functioning of biological neurons as close as possible [1][2], our team is focused on implementing neuromorphic circuits suitable for embedded applications. This objective puts the emphasis on two majors concerns: integration and energy efficiency. In our quest for ultimate integration, we first report on investigating for the best synapse-like technology among the realm of potential candidates. We then report our investigations on the feasibility of large crossbars of synapse-like devices and show that there is still a long way ahead. Finally in an effort to tackle the energy problem, we introduce spike based coding for deep neuromorphic architectures and discuss our argument that spike coding combined with memristive synaptic devices could pave the way for future embedded neuromorphic circuits.
- Published
- 2015