FeFETs for Neuromorphic Systems

Neuromorphic engineering represents one of the most promising computing paradigms for overcoming the limitations of the present-day computers in terms of energy efficiency and processing speed. While traditional neuromorphic circuits are based on complementary metal oxide semiconductor (CMOS) transistors and large capacitors, the recently emerging nanoelectronic devices stand out as promising candidates for building the fundamental neuromorphic elements: neurons and synapses. In this chapter, we illustrate how hafnium oxide-based ferroelectric field-effect transistors (FeFETs) can be used to realize both artificial neurons and synapses for spiking neural networks. In particular, the accumulative switching property of FeFETs will be exploited to mimic the integrate-and-fire neuronal functionality, whereas the continuously tunable synaptic weights and the plasticity will be implemented by the partial polarization switching in large-area devices. Finally, the use of FeFETs for deep neural networks will be briefly discussed.

[1]  Byung Chul Jang,et al.  First Demonstration of a Logic-Process Compatible Junctionless Ferroelectric FinFET Synapse for Neuromorphic Applications , 2018, IEEE Electron Device Letters.

[2]  Stefan Slesazeck,et al.  Mimicking biological neurons with a nanoscale ferroelectric transistor. , 2018, Nanoscale.

[3]  A. Faisal,et al.  Noise in the nervous system , 2008, Nature Reviews Neuroscience.

[4]  Gert Cauwenberghs,et al.  Neuromorphic Silicon Neuron Circuits , 2011, Front. Neurosci.

[5]  Anthony N. Burkitt,et al.  A Review of the Integrate-and-fire Neuron Model: I. Homogeneous Synaptic Input , 2006, Biological Cybernetics.

[6]  G. Bi,et al.  Synaptic Modifications in Cultured Hippocampal Neurons: Dependence on Spike Timing, Synaptic Strength, and Postsynaptic Cell Type , 1998, The Journal of Neuroscience.

[7]  Pritish Narayanan,et al.  Neuromorphic computing using non-volatile memory , 2017 .

[8]  J. Grollier,et al.  A ferroelectric memristor. , 2012, Nature materials.

[9]  T. Morie,et al.  Three-terminal ferroelectric synapse device with concurrent learning function for artificial neural networks , 2012 .

[10]  Shimeng Yu,et al.  AC conductance measurement and analysis of the conduction processes in HfOx based resistive switching memory , 2011 .

[11]  H. Ishiwara,et al.  Adaptive-learning neuron integrated circuits using metal-ferroelectric (SrBi/sub 2/Ta/sub 2/O/sub 9/)-semiconductor (MFS) FET's , 1999, IEEE Electron Device Letters.

[12]  Stefan Slesazeck,et al.  Accumulative Polarization Reversal in Nanoscale Ferroelectric Transistors. , 2018, ACS applied materials & interfaces.

[13]  Vivienne Sze,et al.  Efficient Processing of Deep Neural Networks: A Tutorial and Survey , 2017, Proceedings of the IEEE.

[14]  C Koch,et al.  Complexity and the nervous system. , 1999, Science.

[15]  S. Slesazeck,et al.  Switching Kinetics in Nanoscale Hafnium Oxide Based Ferroelectric Field-Effect Transistors. , 2017, ACS applied materials & interfaces.

[16]  L. Abbott,et al.  Synaptic plasticity: taming the beast , 2000, Nature Neuroscience.

[17]  Chiara Bartolozzi,et al.  Neuromorphic Electronic Circuits for Building Autonomous Cognitive Systems , 2014, Proceedings of the IEEE.

[18]  R. Collongues.  Nonstoichiometry in Ternary Oxides and Materials Science , 1993 .

[19]  Wofgang Maas,et al.  Networks of spiking neurons: the third generation of neural network models , 1997 .

[20]  E. Adrian,et al.  The impulses produced by sensory nerve endings , 1926, The Journal of physiology.

[21]  Nicolas Locatelli,et al.  Learning through ferroelectric domain dynamics in solid-state synapses , 2017, Nature Communications.

[22]  Yukihiro Kaneko,et al.  Supervised Learning Using Spike-Timing-Dependent Plasticity of Memristive Synapses , 2015, IEEE Transactions on Neural Networks and Learning Systems.