1. From Memory in our Brain to Emerging Resistive Memories in Neuromorphic Systems
- Author
-
Olivier Bichler, L. Perniola, Daniele Garbin, E. Vianello, B. DeSalvo, Commissariat à l'énergie atomique et aux énergies alternatives - Laboratoire d'Electronique et de Technologie de l'Information (CEA-LETI), 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), Département d'Architectures, Conception et Logiciels Embarqués-LIST (DACLE-LIST), Laboratoire d'Intégration des Systèmes et des Technologies (LIST (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)), and Laboratoire d'Intégration des Systèmes et des Technologies (LIST)
- Subjects
ReRAM ,Computer science ,Programmable metallization cell ,Complex networks ,artificial synapses ,[SPI]Engineering Sciences [physics] ,Electronic engineering ,cognitive phenomena ,Spiking neural network ,Resistive touchscreen ,Stochastic systems ,Data reduction ,Feedforward neural networks ,Cognitive systems ,Parallel processing systems ,Feed forward ,Brain ,Cognition ,Data handling ,Resistive random-access memory ,Human memory ,Flip flop circuits ,Energy efficiency ,Neuromorphic engineering ,Computer architecture ,Neuromorphic circuits ,Neural networks ,Random access storage ,Efficient energy use - Abstract
Conference of 2015 7th IEEE International Memory Workshop, IMW 2015 ; Conference Date: 17 May 2015 Through 20 May 2015; Conference Code:113230; International audience; In this work, we will focus on the role that new nonvolatile resistive memory technologies (as OxRAM and CBRAM) can play in emerging fields of application, such as neuromorphic circuits, to save energy and increase performance. We will present large-scale energy efficient neuromorphic systems based on ReRAM as stochastic-binary synapses. Prototype applications such as complex visual- and auditory-pattern extraction will be discussed using feedforward spiking neural networks. A parallel will be drawn between these systems and human memory, as recent discoveries on the human brain and cognitive processes may bring benefits and open new perspectives for intelligent data processing.
- Published
- 2015