TY - GEN
T1 - Biologically Plausible Ferroelectric Quasi-Leaky Integrate and Fire Neuron
AU - Dutta, S.
AU - Saha, A.
AU - Panda, P.
AU - Chakraborty, W.
AU - Gomez, J.
AU - Khanna, A.
AU - Gupta, S.
AU - Roy, K.
AU - Datta, S.
N1 - Funding Information:
This work was supported in part by the Semiconductor Research Corporation (SRC) and DARPA.
Funding Information:
Acknowledgement: This work was supported in part by the Semiconductor Research Corporation (SRC) and DARPA. References: [1] P. Diehl Front. Neuro. 2015. [2] A. Saha arxiv.org:1901.07121
Publisher Copyright:
© 2019 The Japan Society of Applied Physics.
PY - 2019/6
Y1 - 2019/6
N2 - Biologically plausible mechanism like homeostasis compliments Hebbian learning to allow unsupervised learning in spiking neural networks [1]. In this work, we propose a novel ferroelectric-based quasi-LIF neuron that induces intrinsic homeostasis. We experimentally characterize and perform phase-field simulations to delineate the non-trivial transient polarization relaxation mechanism associated with multi-domain interaction in poly-crystalline ferroelectric, such as Zr doped HfO2, that underlines the Q-LIF behavior. Network level simulations with the Q-LIF neuron model exhibits a 2.3x reduction in firing rate compared to traditional LIF neuron while maintaining iso-accuracy of 84-85% across varying network sizes. Such an energy-efficient hardware for spiking neuron can enable ultra-low power data processing in energy constrained environments suitable for edge-intelligence.
AB - Biologically plausible mechanism like homeostasis compliments Hebbian learning to allow unsupervised learning in spiking neural networks [1]. In this work, we propose a novel ferroelectric-based quasi-LIF neuron that induces intrinsic homeostasis. We experimentally characterize and perform phase-field simulations to delineate the non-trivial transient polarization relaxation mechanism associated with multi-domain interaction in poly-crystalline ferroelectric, such as Zr doped HfO2, that underlines the Q-LIF behavior. Network level simulations with the Q-LIF neuron model exhibits a 2.3x reduction in firing rate compared to traditional LIF neuron while maintaining iso-accuracy of 84-85% across varying network sizes. Such an energy-efficient hardware for spiking neuron can enable ultra-low power data processing in energy constrained environments suitable for edge-intelligence.
UR - http://www.scopus.com/inward/record.url?scp=85070264708&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85070264708&partnerID=8YFLogxK
U2 - 10.23919/VLSIT.2019.8776487
DO - 10.23919/VLSIT.2019.8776487
M3 - Conference contribution
AN - SCOPUS:85070264708
T3 - Digest of Technical Papers - Symposium on VLSI Technology
SP - T140-T141
BT - 2019 Symposium on VLSI Technology, VLSI Technology 2019 - Digest of Technical Papers
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 39th Symposium on VLSI Technology, VLSI Technology 2019
Y2 - 9 June 2019 through 14 June 2019
ER -