TY - JOUR
T1 - Stochastic Spiking Neural Networks Enabled by Magnetic Tunnel Junctions
T2 - From Nontelegraphic to Telegraphic Switching Regimes
AU - Liyanagedera, Chamika M.
AU - Sengupta, Abhronil
AU - Jaiswal, Akhilesh
AU - Roy, Kaushik
N1 - Publisher Copyright:
© 2017 American Physical Society.
PY - 2017/12/15
Y1 - 2017/12/15
N2 - Stochastic spiking neural networks based on nanoelectronic spin devices can be a possible pathway to achieving "brainlike" compact and energy-efficient cognitive intelligence. The computational model attempt to exploit the intrinsic device stochasticity of nanoelectronic synaptic or neural components to perform learning or inference. However, there has been limited analysis on the scaling effect of stochastic spin devices and its impact on the operation of such stochastic networks at the system level. This work attempts to explore the design space and analyze the performance of nanomagnet-based stochastic neuromorphic computing architectures for magnets with different barrier heights. We illustrate how the underlying network architecture must be modified to account for the random telegraphic switching behavior displayed by magnets with low barrier heights as they are scaled into the superparamagnetic regime. We perform a device-to-system-level analysis on a deep neural-network architecture for a digit-recognition problem on the MNIST data set.
AB - Stochastic spiking neural networks based on nanoelectronic spin devices can be a possible pathway to achieving "brainlike" compact and energy-efficient cognitive intelligence. The computational model attempt to exploit the intrinsic device stochasticity of nanoelectronic synaptic or neural components to perform learning or inference. However, there has been limited analysis on the scaling effect of stochastic spin devices and its impact on the operation of such stochastic networks at the system level. This work attempts to explore the design space and analyze the performance of nanomagnet-based stochastic neuromorphic computing architectures for magnets with different barrier heights. We illustrate how the underlying network architecture must be modified to account for the random telegraphic switching behavior displayed by magnets with low barrier heights as they are scaled into the superparamagnetic regime. We perform a device-to-system-level analysis on a deep neural-network architecture for a digit-recognition problem on the MNIST data set.
UR - http://www.scopus.com/inward/record.url?scp=85038435183&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85038435183&partnerID=8YFLogxK
U2 - 10.1103/PhysRevApplied.8.064017
DO - 10.1103/PhysRevApplied.8.064017
M3 - Article
AN - SCOPUS:85038435183
SN - 2331-7019
VL - 8
JO - Physical Review Applied
JF - Physical Review Applied
IS - 6
M1 - 064017
ER -