Oxide Neuron Devices and Their Applications in Artificial Neural Networks

doi:10.15541/jim20230405

Abstract

Abstract:

Nowadays, artificial intelligence (AI) is playing an increasingly important role in human society. Running AI algorithms represented by deep learning places great demands on computational power of hardware. However, with Moore's Law approaching physical limitations, the traditional Von Neumann computing architecture cannot meet the urgent demand for promoting hardware computational power. The brain-inspired neuromorphic computing (NC) employing an integrated processing-memory architecture is expected to provide an important hardware basis for developing novel AI technologies with low energy consumption and high computational power. Under this conception, artificial neurons and synapses, as the core components of NC systems, have become a research hotspot. This paper aims to provide a comprehensive review on the development of oxide neuron devices. Firstly, several mathematical models of neurons are described. Then, recent progress of Hodgkin-Huxley neurons, leaky integrate-and-fire neurons and oscillatory neurons based on oxide electronic devices is introduced in detail. The effects of device structures and working mechanisms on neuronal performance are systematically analyzed. Next, the hardware implementation of spiking neural networks and oscillatory neural networks based on oxide artificial neurons is demonstrated. Finally, the challenges of oxide neuron devices, arrays and networks, as well as prospect for their applications are pointed out.

Key words: oxide, neuron device, brain-inspired computing, neuromorphic computing, artificial neural network, review

CLC Number:

O472

LI Zongxiao, HU Lingxiang, WANG Jingrui, ZHUGE Fei. Oxide Neuron Devices and Their Applications in Artificial Neural Networks[J]. Journal of Inorganic Materials, 2024, 39(4): 345-358.

Figures/Tables 9

Fig. 1 Typical structure and dynamic response of biological neuron (a) Typical structure of biological neuron; (b) Schematic diagram of voltage-gated ion channels in neural membrane; (c) Dynamic changes of membrane potential under stimulation

Fig. 2 Hodgkin-Huxley (HH) neuron based on VO2 memristors[32] (a) HH neuron circuit based on VO2-based memristors; (b) Output resonator spiking of VO2-based HH neuron; (c) Output inhibition-induced spiking of VO2-based HH neuron; (d) Spike frequency adaptation property of HH neuron

Fig. 3 Leaky integrate-and-fire (LIF) neuron based on antiferroelectric field effect transistor[43] (a) Volatile typical transfer curves of an antiferroelectric field effect transistor (AFeFET) (up) and continuous firing events of an AFeFET neuron under voltage pulses (down); (b) Dynamic of leaky and integration process of an AFeFET neuron under gate pulse with different amplitudes; (c) Artificial neuron circuit diagram based on AFeFET

Fig. 4 LIF neuron based on an antiferromagnetic spintronic device[51] (a) Schematic of an antiferromagnetic spintronic device; (b) Schematic of a polar magneto-optic Kerr effect microscope setup for in-situ magneto-electrical transport probing (up) and its measured domain wall position of hall bar under current stimuli (down); (c) Domain wall position signal (up), neural threshold signal (middle) and output voltage spike dynamics (down) of the antiferromagnetic spintronic device under current stimuli with inset presenting dynamics of domain wall motion

Fig. 5 Researches of LIF neurons based on oxide memristors (a) Typical I-V curve of a NbOx-based TS memristor[63]; (b) Schematic diagram of artificial spiking neuron circuit based on NbOx-based memristor[63]; (c) Oscillation and output spiking characteristics of memristive neuron under constant voltage stimuli[63]; (d) Schematic illustration of an optoelectronic neuron with ITO/IGZO/Ag/Ta2O5/ITO structure[73]; (e) Comparison of fire dynamics of the optoelectronic neuron under dark and ultraviolet light[73]; (f) Fire frequency of the optoelectronic neuron as a function of light intensity at different wavelengths[73]

Fig. 6 Coupling circuits of oscillation neurons and the output mutual waves (a) Circuit schematic of three spin-torque nano-oscillators connected electrically[108]; (b) Unsynchronized oscillatory wave of the three coupled oscillators[108]; (c) Synchronized oscillatory wave of the three coupled oscillators[108]; (d) Coupled circuit consisting of two VO2-based oscillators[96]; (e) In-phase voltage oscillatory waves of VO2-based coupled circuit[96]; (f) Out-of-phase voltage oscillatory waves of VO2-based coupled circuit[96]

Table 1 Performance comparison of HH, LIF and oscillation neurons based on oxides

Type	Device structure	Physics	Auxiliary circuit	Operation stimulus	Highest output frequency	Energy consumption per spike	Advanced function	Ref.
HH	Pt/VO₂/Pt	Mott	2S2R2C* 2S1R3C* 2S2R3C*	Current/ Voltage	<60 kHz	5.6 fJ	23 types of biological neuronal behaviors	[32]
HH	W/WO₃/PEDOT:PSS/Pt	Proton migration	CMOS	2 V	—	—	Local graded potential, all or nothing	[34]
LIF	Si:HfO₂-based FeFET	Polarization switching	6T*	2.4 V	—	—	Integration of excitatory and inhibitory inputs	[42]
	Hf_0.5Zr_0.5O₂-based FeFET	Polarization switching	5T1C*	1.8 V	—	—	Spike frequency adaptation	[44]
	Hf_0.2Zr_0.8O₂-based FeFET	Polarization switching	6T1R*	1.8 V	—	37 fJ	Adjustable output frequency	[43]
	MTJ	Spin	1T*	—	17 MHz	486 fJ	Adjustable output frequency	[51]
	Pt/Ag/TiN/HfAlO_x/Pt	Filament	2R1C*	1.5 V	—	16 fJ	Adjustable output frequency	[111]
	Ag/SiO₂/SiO_2.03/Pt	Filament	2R1C*	0.1 V	—	2 fJ	Adjustable output frequency	[79]
	Au/VO₂/Au	Mott	2R1C*	5 V	1 MHz	2.9 nJ	Adjustable output frequency	[112]
	Si/NbO₂/TiN	Mott	1R*	2 V	900 kHz	38 pJ	Self-protection	[110]
Oscillation	Pt/TaO_x/Ta/Pt	Filament	1R1C*	4-6V	250 MHz	300 μW	Adjustable output frequency	[105]
	Ag/HfO_x/Pt	Filament	1R*	0.6 V	~80 kHz	1.8 µW	Adjustable output frequency	[113]
	Pt/NbO_x/Pt	Mott	1R1C*	4 V	33 MHz	—	Adjustable output frequency	[94]
	VO₂	Mott	1R1C*	2.5 V	1 MHz	735 mW	Coupling	[114]

Fig. 7 Researches on hardware implementations of spiking neural network (SNN) based on oxide neurons (a) Schematic of an SNN with 8×3 array network for unsupervised learning[116]; (b) Evolution of input voltages, neuron currents and synaptic weights in the unsupervised learning process[116]; (c) Neuromorphic circuit based on hybrid memristor/CMOS neurons[119]; (d) Circuit diagram of the V/VOx/HWOx/Pt-based SNN hardware system[120]; (e) Output spike frequency (Vout) as a function of resistance of R-mode device[120]

Fig. 8 VO2 oscillator-based oscillatory neural network (ONN) for Ising Hamiltonian solver[121] (a) Eight-node Ising model; (b) Schematic of a phase-transition nano-oscillator consisting of a VO2-based memristor in series with a transistor (left) and the scanning electron microscopy image of the VO2-based memristor (right); (c) Measured oscillatory waveforms in no-synchronization state, first-harmonic injection-locking and second-harmonic injection-locking

References 130

[1]	ROY K, JAISWAL A, PANDA P. Towards spike-based machine intelligence with neuromorphic computing. Nature, 2019, 575(7784): 607. DOI
[2]	BOAHEN K. Dendrocentric learning for synthetic intelligence. Nature, 2022, 612(7938): 43. DOI
[3]	MEROLLA P A, ARTHUR J V, ALVAREZ-ICAZA R, et al. A million spiking-neuron integrated circuit with a scalable communication network and interface. Science, 2014, 345(6197): 668. DOI URL
[4]	IMAM N, CLELAND T A. Rapid online learning and robust recall in a neuromorphic olfactory circuit. Nature Machine Intelligence, 2020, 2(3): 181. DOI
[5]	PEI J, DENG L, SONG S, et al. Towards artificial general intelligence with hybrid Tianjic chip architecture. Nature, 2019, 572(7767): 106. DOI
[6]	SHEN J, MA D, GU Z, et al. Darwin: a neuromorphic hardware co-processor based on spiking neural networks. Science China Information Sciences, 2015, 59(2): 023401.
[7]	HU L, YANG J, WANG J, et al. All-optically controlled memristor for optoelectronic neuromorphic computing. Advanced Functional Materials, 2021, 31(4): 2005582. DOI URL
[8]	WANG T, RAN H, ZHOU G. Defect-induced analogue resistive switching behavior in FeO_x-based memristor and synaptic paired-pulse facilitation feature. Journal of Inorganic Materials, 2023, 38(4): 437. DOI URL
[9]	WAN C, HE Y, JIANG S, et al. Emerging devices for biologically accurate neuron. ACS Applied Electronic Materials, 2020, 2(2): 389. DOI URL
[10]	FARRANT M, NUSSER Z. Variations on an inhibitory theme: phasic and tonic activation of GABA(A) receptors. Nature Reviews Neuroscience, 2005, 6(3): 215. DOI PMID
[11]	MARKRAM H, TOLEDO-RODRIGUEZ M, WANG Y, et al. Interneurons of the neocortical inhibitory system. Nature Reviews Neuroscience, 2004, 5(10): 793. DOI PMID
[12]	IZHIKEVICH E M. Which model to use for cortical spiking neurons? IEEE Transactions on Neural Networks, 2004, 15(5): 1063. DOI PMID
[13]	HODGKIN A L, HUXLEY A F. The components of membrane conductance in the giant axon of loligo. The Journal of Physiology, 1952, 116(4): 473. DOI URL
[14]	HODGKIN A L, HUXLEY A F. Currents carried by sodium and potassium ions through the membrane of the giant axon of loligo. The Journal of Physiology, 1952, 116(4): 449. DOI URL
[15]	BUGMANN G. Summation and multiplication: two distinct operation domains of leaky integrate-and-fire neurons. Network: Computation in Neural Systems, 1991, 2(4): 489. DOI URL
[16]	XU L F, LI C D, CHEN L. Contrastive analysis of neuron model. Acta Physica Sinica, 2016, 65(24): 240701. DOI URL
[17]	NEUMANN J V. Non-linear capacitance or inductance switching, amplifying and memory organs. USA, 2,815,488, 1957-3-12.
[18]	WIGINGTON R L. A new concept in computing. Proceedings of the IRE, 1959, 47(4): 516. DOI URL
[19]	SOBEL E C, TANK D W. In vivo Ca²⁺ dynamics in a cricket auditory neuron: an example of chemical computation. Science, 1994, 263(5148): 823. DOI URL
[20]	PERON S, GABBIANI F. Spike frequency adaptation mediates looming stimulus selectivity in a collision-detecting neuron. Nature Neuroscience, 2009, 12(3): 318. DOI PMID
[21]	ISAACSON J S, SCANZIANI M. How inhibition shapes cortical activity. Neuron, 2011, 72(2): 231. DOI PMID
[22]	ZHOU Y, CHEN H, WANG Y. Role of lateral inhibition on visual number sense. Frontiers in Computational Neuroscience, 2022, 16: 810448. DOI URL
[23]	MA Q, HAIDER M R, SHRESTHA V L, et al. Bursting Hodgkin-Huxley model-based ultra-low-power neuromimetic silicon neuron. Analog Integrated Circuits and Signal Processing, 2012, 73(1): 329. DOI URL
[24]	KWON M W, BAEK M H, HWANG S, et al. Integrate-and-fire neuron circuit using positive feedback field effect transistor for low power operation. Journal of Applied Physics, 2018, 124(15): 152107. DOI URL
[25]	CHATTERJEE D, KOTTANTHARAYIL A. A CMOS compatible bulk FinFET-based ultra low energy leaky integrate and fire neuron for spiking neural networks. IEEE Electron Device Letters, 2019, 40(8): 1301. DOI URL
[26]	VALASEK J. Piezo-electric and allied phenomena in Rochelle salt. Physical Review, 1921, 17(4): 475. DOI URL
[27]	ISHIBASHI Y, TAKAGI Y. Note on ferroelectric domain switching. Journal of the Physical Society of Japan, 1971, 31(2): 506. DOI URL
[28]	GROLLIER J, QUERLIOZ D, CAMSARI K Y, et al. Neuromorphic spintronics. Nature Electronics, 2020, 3(7): 360. DOI
[29]	CHUA L. Memristor-the missing circuit element. IEEE Transactions on Circuit Theory, 1971, 18(5): 507. DOI URL
[30]	CHUA L. Memristor, Hodgkin-Huxley, and edge of chaos. Nanotechnology, 2013, 24(38): 383001. DOI URL
[31]	PICKETT M D, MEDEIROS-RIBEIRO G, WILLIAMS R S. A scalable neuristor built with mott memristors. Nature Materials, 2013, 12(2): 114. DOI PMID
[32]	YI W, TSANG K K, LAM S K, et al. Biological plausibility and stochasticity in scalable VO₂ active memristor neurons. Nature Communications, 2018, 9: 4661. DOI
[33]	XU Y, GAO S, LI Z, et al. Adaptive Hodgkin-Huxley neuron for retina-inspired perception. Advanced Intelligent Systems, 2022, 4(12): 2200210. DOI URL
[34]	HUANG H M, YANG R, TAN Z H, et al. Quasi-Hodgkin-Huxley neurons with leaky integrate-and-fire functions physically realized with memristive devices. Advanced Materials, 2019, 31(3): 1803849. DOI URL
[35]	PICKETT M D, STANLEY WILLIAMS R. Phase transitions enable computational universality in neuristor-based cellular automata. Nanotechnology, 2013, 24(38): 384002. DOI URL
[36]	BUSCAGLIA M T, BUSCAGLIA V, VIVIANI M, et al. Ferroelectric properties of dense nanocrystalline BaTiO₃ ceramics. Nanotechnology, 2004, 15(9): 1113. DOI URL
[37]	LI Z X, LIU X L, CHEN W J, et al. Switchable diode effect in ferroelectric thin film: high dependence on poling process and temperature. AIP Advances, 2014, 4(12): 127111. DOI URL
[38]	WANG J, ZHENG H, MA Z, et al. Epitaxial BiFeO₃ thin films on Si. Applied Physics Letters, 2004, 85(13): 2574. DOI URL
[39]	BÖSCKE T S, MÜLLER J, BRÄUHAUS D, et al. Ferroelectricity in hafnium oxide thin films. Applied Physics Letters, 2011, 99(10): 102903. DOI URL
[40]	MULAOSMANOVIC H, MIKOLAJICK T, SLESAZECK S. Accumulative polarization reversal in nanoscale ferroelectric transistors. ACS Applied Materials & Interfaces, 2018, 10(28): 23997.
[41]	MULAOSMANOVIC H, MIKOLAJICK T, SLESAZECK S. Random number generation based on ferroelectric switching. IEEE Electron Device Letters, 2018, 39(1): 135. DOI URL
[42]	MULAOSMANOVIC H, CHICCA E, BERTELE M, et al. Mimicking biological neurons with a nanoscale ferroelectric transistor. Nanoscale, 2018, 10(46): 21755. DOI PMID
[43]	CAO R, ZHANG X, LIU S, et al. Compact artificial neuron based on anti-ferroelectric transistor. Nature Communications, 2022, 13: 7018. DOI PMID
[44]	CHEN C, YANG M, LIU S, et al. Bio-inspired Neurons Based on Novel Leaky-FeFET with Ultra-low Hardware Cost and Advanced Functionality for All-ferroelectric Neural NetWork. Symposium on VLSI Technology, Kyoto, 2019: T136-T137.
[45]	LUO J, YU L, LIU T, et al. Capacitor-less Stochastic Leaky-FeFET Neuron of Both Excitatory and Inhibitory Connections for SNN with Reduced Hardware Cost. IEEE International Electron Devices Meeting, San Francisco, 2019: 6.4.1-6.4.4.
[46]	KATINE J A, ALBERT F J, BUHRMAN R A, et al. Current- driven magnetization reversal and spin-wave excitations in Co/Cu/Co pillars. Physical Review Letters, 2000, 84(14): 3149. DOI URL
[47]	TSOI M, JANSEN A G, BASS J, et al. Generation and detection of phase-coherent current-driven magnons in magnetic multilayers. Nature, 2000, 406(6791): 46. DOI
[48]	KURENKOV A, DUTTAGUPTA S, ZHANG C, et al. Artificial neuron and synapse realized in an antiferromagnet/ferromagnet heterostructure using dynamics of spin-orbit torque switching. Advanced Materials, 2019, 31(23): 1900636. DOI URL
[49]	YANG S, SHIN J, KIM T, et al. Integrated neuromorphic computing networks by artificial spin synapses and spin neurons. NPG Asia Materials, 2021, 13(1): 11. DOI
[50]	LIU J H, XU T, FENG H M, et al. Compensated ferrimagnet based artificial synapse and neuron for ultrafast neuromorphic computing. Advanced Functional Materials, 2022, 32(1): 2107870. DOI URL
[51]	WANG D, TANG R, LIN H, et al. Spintronic leaky-integrate-fire spiking neurons with self-reset and winner-takes-all for neuromorphic computing. Nature Communications, 2023, 14: 1068. DOI PMID
[52]	MA W J, LIN S P, LUO J M, et al. Highly uniform bipolar resistive switching characteristics in TiO₂/BaTiO₃/TiO₂ multilayer. Applied Physics Letters, 2013, 103(26): 262903. DOI URL
[53]	ZHOU G, SUN B, HU X, et al. Negative photoconductance effect: an extension function of the TiO_x-based memristor. Advanced Science, 2021, 8(13): 2003765. DOI URL
[54]	ZHUGE F, LI K, FU B, et al. Mechanism for resistive switching in chalcogenide-based electrochemical metallization memory cells. AIP Advances, 2015, 5(5): 057125. DOI URL
[55]	HU L, FU S, CHEN Y, et al. Ultrasensitive memristive synapses based on lightly oxidized sulfide films. Advanced Materials, 2017, 29(24): 1606927. DOI URL
[56]	ZHANG S R, ZHOU L, MAO J Y, et al. Artificial synapse emulated by charge trapping-based resistive switching device. Advanced Materials Technologies, 2019, 4(2): 1800342. DOI URL
[57]	YOU J, YANG D, LI C, et al. Double dielectric layer metal-oxide memristor: design and applications. Journal of Inorganic Materials, 2023, 38(4): 387. DOI
[58]	ZHUGE X, ZHU R, WANG J, et al. Oxide memristors for brain-inspired computing. Journal of Inorganic Materials, 2023, 38(10): 1149. DOI URL
[59]	YANG J, HU L, SHEN L, et al. Optically driven intelligent computing with ZnO memristor. Fundamental Research, 2024, 4(1): 158. DOI URL
[60]	LASHKARE S, CHOUHAN S, CHAVAN T, et al. PCMO RRAM for integrate-and-fire neuron in spiking neural networks. IEEE Electron Device Letters, 2018, 39(4): 484. DOI URL
[61]	WANG J J, HU S G, ZHAN X T, et al. Handwritten-digit recognition by hybrid convolutional neural network based on HfO₂ memristive spiking-neuron. Scientific Reports, 2018, 8: 12546. DOI
[62]	TUMA T, PANTAZI A, LE GALLO M, et al. Stochastic phase-change neurons. Nature Nanotechnology, 2016, 11(8): 693. DOI PMID
[63]	ZHU J X, ZHANG X M, WANG R, et al. Flexible memristive spiking neuron for neuromorphic sensing and computing. Acta Physica Sinica, 2022, 71(14): 148503. DOI URL
[64]	GU Y N, LIANG Y, WANG G Y, et al. Design of NbO_x memristive neuron and its application in spiking neural networks. Acta Physica Sinica, 2022, 71(11): 110501. DOI URL
[65]	STOLIAR P, TRANCHANT J, CORRAZE B, et al. A leaky- integrate-and-fire neuron analog realized with a Mott insulator. Advanced Functional Materials, 2017, 27(11): 1604740. DOI URL
[66]	ADDA C, CORRAZE B, STOLIAR P, et al. Mott insulators: a large class of materials for leaky integrate and fire (LIF) artificial neuron. Journal of Applied Physics, 2018, 124(15): 152124. DOI URL
[67]	PARIHAR A, JERRY M, DATTA S, et al. Stochastic IMT (insulator-metal-transition) neurons: an interplay of thermal and threshold noise at bifurcation. Frontiers in Neuroscience, 2018, 12: 210. DOI PMID
[68]	CHEN P, ZHANG X M, WU Z H, et al. High-yield and uniform NbO_x-based threshold switching devices for neuron applications. IEEE Transactions on Electron Devices, 2022, 69(5): 2391. DOI URL
[69]	SUN Y M, SONG C, YIN S Q, et al. Design of a controllable redox-diffusive threshold switching memristor. Advanced Electronic Materials, 2020, 6(11): 2000695. DOI URL
[70]	WANG Z, JOSHI S, SAVEL'EV S E, et al. Memristors with diffusive dynamics as synaptic emulators for neuromorphic computing. Nature Materials, 2017, 16(1): 101. DOI PMID
[71]	ZHANG X M, WANG W, LIU Q, et al. An artificial neuron based on a threshold switching memristor. IEEE Electron Device Letters, 2018, 39(2): 308. DOI URL
[72]	LEE D, KWAK M, MOON K, et al. Various threshold switching devices for integrate and fire neuron applications. Advanced Electronic Materials, 2019, 5(9): 1800866. DOI URL
[73]	WANG X, CHEN C, ZHU L, et al. Vertically integrated spiking cone photoreceptor arrays for color perception. Nature Communications, 2023, 14: 3444. DOI PMID
[74]	WANG Y, XU H, WANG W, et al. A configurable artificial neuron based on a threshold-tunable TiN/NbOₓ/Pt memristor. IEEE Electron Device Letters, 2022, 43(4): 631. DOI URL
[75]	SPECHT D F. Probabilistic neural networks and the polynomial adaline as complementary techniques for classification. IEEE Transactions on Neural Networks, 1990, 1(1): 111. PMID
[76]	LI Z X, GENG X Y, WANG J, et al. Emerging artificial neuron devices for probabilistic computing. Frontiers in Neuroscience, 2021, 15: 717947. DOI URL
[77]	CARBONI R, IELMINI D. Stochastic memory devices for security and computing. Advanced Electronic Materials, 2019, 5(9): 1900198. DOI URL
[78]	WANG K, HU Q, GAO B, et al. Threshold switching memristor-based stochastic neurons for probabilistic computing. Materials Horizons, 2021, 8(2): 619. DOI PMID
[79]	BOUSOULAS P, TSIOUSTAS C, HADFIELD J, et al. Low power stochastic neurons from SiO₂-based bilayer conductive bridge memristors for probabilistic spiking neural network applications-part I: experimental characterization. IEEE Transactions on Electron Devices, 2022, 69(5): 2360. DOI URL
[80]	BOUSOULAS P, TSIOUSTAS C, HADFIELD J, et al. Low power stochastic neurons from SiO₂-based bilayer conductive bridge memristors for probabilistic spiking neural network applications-part II: modeling. IEEE Transactions on Electron Devices, 2022, 69(5): 2368. DOI URL
[81]	CHOI S, KIM G S, YANG J, et al. Controllable SiO_x nanorod memristive neuron for probabilistic bayesian inference. Advanced Materials, 2022, 34(1): 2104598. DOI URL
[82]	DANG B J, LIU K Q, ZHU J D, et al. Stochastic neuron based on IGZO Schottky diodes for neuromorphic computing. APL Materials, 2019, 7(7): 071114. DOI URL
[83]	MAO H W, HE Y L, CHEN C S, et al. A spiking stochastic neuron based on stacked InGaZnO memristors. Advanced Electronic Materials, 2022, 8(2): 2100918. DOI URL
[84]	YAN L, PEI Y F, WANG J J, et al. High-speed Si films based threshold switching device and its artificial neuron application. Applied Physics Letters, 2021, 119(15): 153507. DOI URL
[85]	WANG J, TENG C, ZHANG Z, et al. A scalable artificial neuron based on ultrathin two-dimensional titanium oxide. ACS Nano, 2021, 15(9): 15123. DOI PMID
[86]	CHEN Y H, WANG Y, LUO Y H, et al. Realization of artificial neuron using MXene bi-directional threshold switching memristors. IEEE Electron Device Letters, 2019, 40(10): 1686. DOI URL
[87]	KIM T, KIM S H, PARK J H, et al. An artificial neuron using a bipolar electrochemical metallization switch and its enhanced spiking properties through filament confinement. Advanced Electronic Materials, 2021, 7(1): 2000410. DOI URL
[88]	TIAN Y, ZHU X, SUN C, et al. Intrinsically stretchable threshold switching memristor for artificial neuron implementations. Journal of Inorganic Materials, 2023, 38(4): 413. DOI
[89]	LI H, HU J, CHEN A, et al. Single-transistor neuron with excitatory-inhibitory spatiotemporal dynamics applied for neuronal oscillations. Advanced Materials, 2022, 34(51): 2207371. DOI URL
[90]	CHEN C, HE Y, MAO H, et al. A photoelectric spiking neuron for visual depth perception. Advanced Materials, 2022, 34(20): 2201895. DOI URL
[91]	WU Q, DANG B, LU C, et al. Spike encoding with optic sensory neurons enable a pulse coupled neural network for ultraviolet image segmentation. Nano Letters, 2020, 20(11): 8015. DOI PMID
[92]	PEI Y, YAN L, WU Z, et al. Artificial visual perception nervous system based on low-dimensional material photoelectric memristors. ACS Nano, 2021, 15(11): 17319. DOI PMID
[93]	YOON S M, TOKUMITSU E, ISHIWARA H. Ferroelectric neuron integrated circuits using SrBi₂Ta₂O₉-gate FET's and CMOS Schmitt-trigger oscillators. IEEE Transactions on Electron Devices, 2000, 47(8): 1630. DOI URL
[94]	GAO L G, CHEN P Y, YU S M. NbO_x based oscillation neuron for neuromorphic computing. Applied Physics Letters, 2017, 111(10): 103503. DOI URL
[95]	ZHAO B, RAVICHANDRAN J. Low-power microwave relaxation oscillators based on phase-change oxides for neuromorphic computing. Physical Review Applied, 2019, 11(1): 014020. DOI URL
[96]	CORTI E, GOTSMANN B, MOSELUND K, et al. Scaled resistively-coupled VO₂ oscillators for neuromorphic computing. Solid-State Electronics, 2020, 168: 107729. DOI URL
[97]	LIU H, WU T, YAN X, et al. A tantalum disulfide charge-density- wave stochastic artificial neuron for emulating neural statistical properties. Nano Letters, 2021, 21(8): 3465. DOI URL
[98]	WU L D, WANG Z W, BAO L, et al. Implementation of neuronal intrinsic plasticity by oscillatory device in spiking neural network. IEEE Transactions on Electron Devices, 2022, 69(4): 1830. DOI URL
[99]	KISELEV S I, SANKEY J C, KRIVOROTOV I N, et al. Microwave oscillations of a nanomagnet driven by a spin-polarized current. Nature, 2003, 425(6956): 380. DOI
[100]	BERTOTTI G, SERPICO C, MAYERGOYZ I D, et al. Magnetization switching and microwave oscillations in nanomagnets driven by spin-polarized currents. Physical Review Letters, 2005, 94(12): 127206. DOI URL
[101]	YOGENDRA K, FAN D, ROY K. Coupled spin torque nano oscillators for low power neural computation. IEEE Transactions on Magnetics, 2015, 51(10): 4003909.
[102]	AZAM M A, BHATTACHARYA D, QUERLIOZ D, et al. Resonate and fire neuron with fixed magnetic skyrmions. Journal of Applied Physics, 2018, 124(15): 152122. DOI URL
[103]	LIANG X, ZHANG X C, XIA J, et al. A spiking neuron constructed by the skyrmion-based spin torque nano-oscillator. Applied Physics Letters, 2020, 116(12): 122402. DOI URL
[104]	TANG L Z, HUANG Y, WANG C, et al. Halide perovskite memristor with ultra-high-speed and robust flexibility for artificial neuron applications. Journal of Materials Chemistry C, 2022, 10(39): 14695. DOI URL
[105]	SHARMA A A, LI Y, SKOWRONSKI M, et al. High-frequency TaO_x-based compact oscillators. IEEE Transactions on Electron Devices, 2015, 62(11): 3857. DOI URL
[106]	YU Y, ZHAO B, GOODWILL J M, et al. Electrical and thermal dynamics of self-oscillations in TaO_x-based threshold switching devices. ACS Applied Electronic Materials, 2020, 2(3): 683. DOI URL
[107]	KAKA S, PUFALL M R, RIPPARD W H, et al. Mutual phase- locking of microwave spin torque nano-oscillators. Nature, 2005, 437(7057): 389. DOI
[108]	ROMERA M, TALATCHIAN P, TSUNEGI S, et al. Binding events through the mutual synchronization of spintronic nano- neurons. Nature Communications, 2022, 13: 883. DOI
[109]	WOO J, WANG P N, YU S M. Integrated crossbar array with resistive synapses and oscillation neurons. IEEE Electron Device Letters, 2019, 40(8): 1313. DOI URL
[110]	ZHANG X, ZHUO Y, LUO Q, et al. An artificial spiking afferent nerve based on Mott memristors for neurorobotics. Nature Communications, 2020, 11: 51. DOI PMID
[111]	LU Y F, LI Y, LI H, et al. Low-power artificial neurons based on Ag/TiN/HfAlO_x/Pt threshold switching memristor for neuromorphic computing. IEEE Electron Device Letters, 2020, 41(8): 1245. DOI URL
[112]	YUAN R, DUAN Q, TIW P J, et al. A calibratable sensory neuron based on epitaxial VO₂ for spike-based neuromorphic multisensory system. Nature Communications, 2022, 13: 3973. DOI
[113]	HUA Q, WU H, GAO B, et al. Low-voltage oscillatory neurons for memristor-based neuromorphic systems. Glob Chall, 2019, 3(11): 1900015.
[114]	SHAMSI J, AVEDILLO M J, LINARES-BARRANCO B, et al. Hardware implementation of differential oscillatory neural networks using VO₂-based oscillators and memristor-bridge circuits. Frontiers in Neuroscience, 2021, 15: 674567. DOI URL
[115]	YANG Y F, LI H L. Tellurium-based artificial neuron: capturing biological complexity while keeping it simple. Advanced Electronic Materials, 2022, 8(9): 2200094. DOI URL
[116]	WANG Z R, JOSHI S, SAVEL'EV S, et al. Fully memristive neural networks for pattern classification with unsupervised learning. Nature Electronics, 2018, 1(2): 137. DOI
[117]	DUAN Q, JING Z, ZOU X, et al. Spiking neurons with spatiotemporal dynamics and gain modulation for monolithically integrated memristive neural networks. Nature Communications, 2020, 11: 3399. DOI PMID
[118]	LI X, TANG J, ZHANG Q, et al. Power-efficient neural network with artificial dendrites. Nature Nanotechnology, 2020, 15(9): 776. DOI PMID
[119]	ZHANG X, LU J, WANG Z, et al. Hybrid memristor-CMOS neurons for in-situ learning in fully hardware memristive spiking neural networks. Science Bulletin, 2021, 66(16): 1624. DOI URL
[120]	FU Y Y, ZHOU Y, HUANG X D, et al. Reconfigurable synaptic and neuronal functions in a V/VO_x/HfWO_x/Pt memristor for nonpolar spiking convolutional neural network. Advanced Functional Materials, 2022, 32(23): 2111996. DOI URL
[121]	DUTTA S, KHANNA A, ASSOA A S, et al. An Ising Hamiltonian solver based on coupled stochastic phase-transition nano- oscillators. Nature Electronics, 2021, 4(7): 502. DOI
[122]	CSABA G, POROD W. Coupled oscillators for computing: a review and perspective. Applied Physics Reviews, 2020, 7(1): 011302. DOI URL
[123]	XIA M, WENG L, WANG Z, et al. Sequence memory based on an oscillatory neural network. Science China Information Sciences, 2014, 57(7): 072203.
[124]	KHITUN A, LIU G, BALANDIN A A. Two-dimensional oscillatory neural network based on room-temperature charge- density-wave devices. IEEE Transactions on Nanotechnology, 2017, 16(5): 860. DOI URL
[125]	VELICHKO A, BELYAEV M, PUTROLAYNEN V, et al. Modeling of thermal coupling in VO₂-based oscillatory neural networks. Solid-State Electronics, 2018, 139: 8. DOI URL
[126]	NUNEZ J, AVEDILLO M J, JIMENEZ M, et al. Oscillatory neural networks using VO₂ based phase encoded logic. Frontiers in Neuroscience, 2021, 15: 655823. DOI URL
[127]	LAUGHLIN S B, DE RUYTER VAN STEVENINCK R R, ANDERSON J C. The metabolic cost of neural information. Nature Neuroscience, 1998, 1(1): 36. DOI PMID
[128]	PULIDO C, RYAN T A. Synaptic vesicle pools are a major hidden resting metabolic burden of nerve terminals. Science Advances, 2021, 7(49): eabi9027. DOI URL
[129]	WU R J, JIA S J, GOTOH T, et al. Screening switching materials with low leakage current and high thermal stability for neuromorphic computing. Advanced Electronic Materials, 2022, 8(9): 2200150. DOI URL
[130]	HAMILTON T J, AFSHAR S, VAN SCHAIK A, et al. Stochastic electronics: a neuro-inspired design paradigm for integrated circuits. Proceedings of the IEEE, 2014, 102(5): 843. DOI URL