TY - JOUR
T1 - Probabilistic Deep Spiking Neural Systems Enabled by Magnetic Tunnel Junction
AU - Sengupta, Abhronil
AU - Parsa, Maryam
AU - Han, Bing
AU - Roy, Kaushik
N1 - Funding Information:
This work was supported in part by the National Science Foundation, in part by Semiconductor Research Corporation, in part by the Center for Spintronic Materials, Interfaces, and Novel Architectures sponsored by Microelectronics Advanced Research Corporation and Defense Advanced Research Projects Agency through the one of six centers of STARnet, in part by the National Security Science and Engineering Faculty Fellowship, and in part by Intel Corporation.
Publisher Copyright:
© 2016 IEEE.
PY - 2016/7
Y1 - 2016/7
N2 - Deep spiking neural networks are becoming increasingly powerful tools for cognitive computing platforms. However, most of the existing studies on such computing models are developed with limited insights on the underlying hardware implementation, resulting in area and power expensive designs. Although several neuromimetic devices emulating neural operations have been proposed recently, their functionality has been limited to very simple neural models that may prove to be inefficient at complex recognition tasks. In this paper, we venture into the relatively unexplored area of utilizing the inherent device stochasticity of such neuromimetic devices to model complex neural functionalities in a probabilistic framework in the time domain. We consider the implementation of a deep spiking neural network capable of performing high-accuracy and low-latency classification tasks, where the neural computing unit is enabled by the stochastic switching behavior of a magnetic tunnel junction. The simulation studies indicate an energy improvement of 20 × over a baseline CMOS design in 45-nm technology.
AB - Deep spiking neural networks are becoming increasingly powerful tools for cognitive computing platforms. However, most of the existing studies on such computing models are developed with limited insights on the underlying hardware implementation, resulting in area and power expensive designs. Although several neuromimetic devices emulating neural operations have been proposed recently, their functionality has been limited to very simple neural models that may prove to be inefficient at complex recognition tasks. In this paper, we venture into the relatively unexplored area of utilizing the inherent device stochasticity of such neuromimetic devices to model complex neural functionalities in a probabilistic framework in the time domain. We consider the implementation of a deep spiking neural network capable of performing high-accuracy and low-latency classification tasks, where the neural computing unit is enabled by the stochastic switching behavior of a magnetic tunnel junction. The simulation studies indicate an energy improvement of 20 × over a baseline CMOS design in 45-nm technology.
UR - http://www.scopus.com/inward/record.url?scp=84971440970&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84971440970&partnerID=8YFLogxK
U2 - 10.1109/TED.2016.2568762
DO - 10.1109/TED.2016.2568762
M3 - Article
AN - SCOPUS:84971440970
SN - 0018-9383
VL - 63
SP - 2963
EP - 2970
JO - IEEE Transactions on Electron Devices
JF - IEEE Transactions on Electron Devices
IS - 7
M1 - 7478636
ER -