Memristive Mixed-Signal Neuromorphic Systems: Energy-Efficient Learning at the Circuit-Level
Open Access
- 23 November 2017
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Emerging and Selected Topics in Circuits and Systems
- Vol. 8 (1), 125-136
- https://doi.org/10.1109/jetcas.2017.2777181
Abstract
Neuromorphic computing is a non-von Neumann computer architecture for the post Moore’s law era of computing. Since a main focus of the post Moore’s law era is energy-efficient computing with fewer resources and less area, neuromorphic computing contributes effectively in this research. In this paper we present a memristive neuromorphic system for improved power and area efficiency. Our particular mixed-signal approach implements neural networks with spiking events in a synchronous way. Moreover, the use of nano-scale memristive devices saves both area and power in the system. We also provide device-level considerations that make the system more energy-efficient. The proposed system additionally includes synchronous digital long term plasticity (DLTP), an online learning methodology that helps the system train the neural networks during the operation phase and improves the efficiency in learning considering the power consumption and area overhead.Keywords
Funding Information
- Air Force Research Laboratory (FA8750-16-1-0065)
- U.S. Department of Energy, Office of Science, Office of Advanced Scientific Computing Research (DE-AC05-00OR22725)
This publication has 41 references indexed in Scilit:
- Ultrafast Synaptic Events in a Chalcogenide MemristorScientific Reports, 2013
- STDP and STDP variations with memristors for spiking neuromorphic learning systemsFrontiers in Neuroscience, 2013
- Lognormal switching times for titanium dioxide bipolar memristors: origin and resolutionNanotechnology, 2011
- Artificial neural networks in hardware: A survey of two decades of progressNeurocomputing, 2010
- Memristor−CMOS Hybrid Integrated Circuits for Reconfigurable LogicNano Letters, 2009
- High-Density Crossbar Arrays Based on a Si Memristive SystemNano Letters, 2009
- Synaptic Modification by Correlated Activity: Hebb's Postulate RevisitedAnnual Review of Neuroscience, 2001
- Networks of spiking neurons: The third generation of neural network modelsNeural Networks, 1997
- A CMOS analog adaptive BAM with on-chip learning and weight refreshingIEEE Transactions on Neural Networks, 1993
- Analog CMOS synaptic learning circuits adapted from invertebrate biologyIEEE Transactions on Circuits and Systems, 1991