A review of memristive reservoir computing for temporal data processing and sensing

Abstract

Reservoir computing (RC) is a promising paradigm for machine learning that uses a fixed, randomly generated network, known as the reservoir, to process input data. A memristor with fading memory and nonlinearity characteristics was adopted as a physical reservoir to implement the hardware RC system. This article reviews the device requirements for effective memristive reservoir implementation and methods for obtaining higher-dimensional reservoirs for improving RC system performance. In addition, recent in-sensor RC system studies, which use a memristor that the resistance is changed by an optical signal to realize an energy-efficient machine vision, are discussed. Finally, the limitations that the memristive and in-sensor RC systems encounter when attempting to improve performance further are discussed, and future directions that may overcome these challenges are suggested.

Keywords

in-sensor reservoir computing / memristive reservoir computing / memristor / reservoir computing

References

[1] Lecun Y, Bengio Y, Hinton G. Deep learning. Nature. 2015; 521(7553): 436-444.

[2] Jain AK, Mao J, Mohiuddin KM. Artificial neural networks: a tutorial. Computer. 1996; 29(3): 31-44.

[3] Schmidhuber J. Deep learning in neural networks: an overview. Neural Network. 2015; 61: 85-117.

[4] Abiodun OI, Jantan A, Omolara AE, Dada KV, Mohamed NA, Arshad H. State-of-the-art in artificial neural network applications: a survey. Heliyon. 2018; 4(11):e00938.

[5] Jaeger H. The ‘echo state’ approach to analysing and training recurrent neural networks. GMD Rep. 2001; 148(34): 13.

[6] Maass WW, Markram H, Natschläger T, Maass WW, Markram H. The “liquid computer”: a novel strategy for real-time computing on time series. Telematik. 2002; 8(1): 39-43.

[7] Lukoševičius M, Jaeger H. Reservoir computing approaches to recurrent neural network training. Comput Sci Rev. 2009; 3: 127-149.

[8] Schrauwen B, Verstraeten D, Van Campenhout J. An overview of reservoir computing: theory, applications and implementations. In:ESANN 2007 Proceedings - 15th European Symposium on Artificial Neural Networks; 2007: 471-482.

[9] Tanaka G, Yamane T, Héroux JB, et al. Recent advances in physical reservoir computing: a review. Neural Network. 2019; 115: 100-123.

[10] Appeltant L, Soriano M, Van der Sande G, et al. Information processing using a single dynamical node as complex system. Nat Commun. 2011; 2(1):468.

[11] Tong Z,Tanaka G. Reservoir computing with untrained convolutional neural networks for image recognition. In: Proceedings - International Conference on Pattern Recognition; 2018.

[12] Jalalvand A, Demuynck K, De Neve W, Martens JP. On the application of reservoir computing networks for noisy image recognition. Neurocomputing. 2018; 277: 237-248.

[13] Jacobson P, Shirao M, Yu K, Su GL, Wu MC. Hybrid convolutional optoelectronic reservoir computing for image recognition. J Light Technol. 2022; 40(3): 692-699.

[14] Jalalvand A,Van Wallendael G, Van De Walle R. Real-time reservoir computing network-based systems for detection tasks on visual contents. In: Proceedings - 7th International Conference on Computational Intelligence, Communication Systems and Networks, CICSyN 2015; 2015.

[15] Canaday D, Griffith A, Gauthier DJ. Rapid time series prediction with a hardware-based reservoir computer. Chaos. 2018; 28(12):123119.

[16] Bianchi FM, Scardapane S, Lokse S, Jenssen R. Reservoir computing approaches for representation and classification of multivariate time series. IEEE Transact Neural Networks Learn Syst. 2021; 32(5): 2169-2179.

[17] Xia Q, Yang JJ. Memristive crossbar arrays for brain-inspired computing. Nat Mater. 2019; 18(4): 309-323.

[18] Kim Y, Kim J, Kim SS, et al. Kernel application of the stacked crossbar array composed of self-rectifying resistive switching memory for convolutional neural networks. Adv Intell Syst. 2020; 2:1900116.

[19] Yao P, Wu H, Gao B, et al. Fully hardware-implemented memristor convolutional neural network. Nature. 2020; 577(7792): 641-646.

[20] Li C, Belkin D, Li Y, et al. Efficient and self-adaptive in-situ learning in multilayer memristor neural networks. Nat Commun. 2018; 9(1):2385.

[21] Jeong DS, Hwang CS. Nonvolatile memory materials for neuromorphic intelligent machines. Adv Mater. 2018; 30(42): 1-27.

[22] Zhong Y, Tang J, Li X, et al. A memristor-based analogue reservoir computing system for real-time and power-efficient signal processing. Nat Electron. 2022; 5(10): 672-681.

[23] Cai F, Correll JM, Lee SH, et al. A fully integrated reprogrammable memristor-CMOS system for efficient multiply-accumulate operations. Nat Electron. 2019; 2(7): 290-299.

[24] Sun L, Wang Z, Jiang J, et al. In-sensor reservoir computing for language learning via two-dimensional memristors. Sci Adv. 2021; 7(20).

[25] Du W, Li C, Huang Y, et al. An optoelectronic reservoir computing for temporal information processing. IEEE Electron Device Lett. 2022; 43(3): 406-409.

[26] Liu K, Zhang T, Dang B, et al. An optoelectronic synapse based on α-In₂Se₃ with controllable temporal dynamics for multimode and multiscale reservoir computing. Nat Electron. 2022; 5(11): 761-773.

[27] Zhang Z, Zhao X, Zhang X, et al. In-sensor reservoir computing system for latent fingerprint recognition with deep ultraviolet photo-synapses and memristor array. Nat Commun. 2022; 13: 1-9.

[28] Sun Y, Li Q, Zhu X, et al. In-sensor reservoir computing based on optoelectronic Synapse. Adv Intell Syst. 2022; 5(1):2200196.

[29] Lao J, Yan M, Tian B, et al. Ultralow-power machine vision with self-powered sensor reservoir. Adv Sci. 2022; 9(15).

[30] Jang YH, Kim W, Kim J, et al. Time-varying data processing with nonvolatile memristor-based temporal kernel. Nat Commun. 2021; 12(1):5727.

[31] Zhong Y, Tang J, Li X, Gao B, Qian H, Wu H. Dynamic memristor-based reservoir computing for high-efficiency temporal signal processing. Nat Commun. 2021; 12(1):408.

[32] Milano G, Pedretti G, Montano K, et al. In materia reservoir computing with a fully memristive architecture based on self-organizing nanowire networks. Nat Mater. 2022; 21(2): 195-202.

[33] Liang X, Luo Y, Pei Y, Wang M, Liu C. Multimode transistors and neural networks based on ion-dynamic capacitance. Nat Electron. 2022; 5(12): 859-869.

[34] Yang Y, Cui H, Ke S, et al. Reservoir computing based on electric-double-layer coupled InGaZnO artificial synapse. Appl Phys Lett. 2023; 122(4).

[35] Sun Y, Li Q, Zhu X, et al. In-Sensor reservoir computing based on optoelectronic synapse. Adv Intell Syst. 2023; 5(1):2200196.

[36] Schaetti N,Salomon M, Couturier R. Echo state networks-based reservoir computing for MNIST handwritten digits recognition. In: Proceedings - 19th IEEE International Conference on Computational Science and Engineering, 14th IEEE International Conference on Embedded and Ubiquitous Computing and 15th International Symposium on Distributed Computing and Applications to Business. Engi; 2017.

[37] Li D, Han M, Wang J. Chaotic time series prediction based on a novel robust echo state network. IEEE Transact Neural Networks Learn Syst. 2012; 23(5): 787-799.

[38] Torrejon J, Riou M, Araujo FA, et al. Neuromorphic computing with nanoscale spintronic oscillators. Nature. 2017; 547(7664): 428-431.

[39] Du C, Cai F, Zidan MA, Ma W, Lee SH, Lu WD. Reservoir computing using dynamic memristors for temporal information processing. Nat Commun. 2017; 8: 1-10.

[40] Maass W, Natschläger T, Markram H. Real-time computing without stable states: a new framework for neural computation based on perturbations. Neural Comput. 2002; 14(11): 2531-2560.

[41] Strukov DB, Snider GS, Stewart DR, Williams RS. The missing memristor found. Nature. 2008; 453(7191): 80-83.

[42] Kim KM, Zhang J, Graves C, et al. Low-power, self-rectifying, and forming-free memristor with an asymmetric programing voltage for a high-density crossbar application. Nano Lett. 2016; 16(11): 6724-6732.

[43] Kim GH, Lee JH, Ahn Y, et al. 32 × 32 crossbar array resistive memory composed of a stacked Schottky diode and unipolar resistive memory. Adv Funct Mater. 2013; 23(11): 1440-1449.

[44] Li C, Hu M, Li Y, et al. Analogue signal and image processing with large memristor crossbars. Nat Electron. 2018; 1: 52-59.

[45] Lee YK, Jeon JW, Park ES, et al. Matrix mapping on crossbar memory arrays with resistive interconnects and its use in in-memory compression of biosignals. Micromachines. 2019; 10(5):306.

[46] Kim Y, Kwon YJ, Kim J, et al. Novel selector-induced current-limiting effect through asymmetry control for high-density one-selector-one-resistor crossbar arrays. Adv Electron Mater. 2019; 5(7): 1-11.

[47] Wang Z, Joshi S, Savel’ev SE, et al. Memristors with diffusive dynamics as synaptic emulators for neuromorphic computing. Nat Mater. 2017; 16(1): 101-108.

[48] Midya R, Wang Z, Asapu S, et al. Reservoir computing using diffusive memristors. Adv Intell Syst. 2019; 1(7):1900084.

[49] Wang R, Yang JQ, Mao JY, et al. Recent advances of volatile memristors: devices, mechanisms, and applications. Adv Intell Syst. 2020; 2(9):2000055.

[50] Chen PY, Peng X, Yu S. NeuroSim+:an integrated device-to-algorithm framework for benchmarking synaptic devices and array architectures. In: Technical Digest - International Electron Devices Meeting, IEDM; 2018.

[51] Moon J, Ma W, Shin JH, et al. Temporal data classification and forecasting using a memristor-based reservoir computing system. Nat Electron. 2019; 2(10): 480-487.

[52] Zhu X, Wang Q, Lu WD. Memristor networks for real-time neural activity analysis. Nat Commun. 2020; 11(1):2439.

[53] Carroll TL, Pecora LM. Network structure effects in reservoir computers. Chaos. 2019; 29(8):083130.

[54] Büsing L, Schrauwen B, Legenstein R. Connectivity, dynamics, and memory in reservoir computing with binary and analog neurons. Neural Comput. 2010; 22(5): 1272-1311.

[55] Lukoševičius M, Jaeger H, Schrauwen B. Reservoir computing trends. Kunstliche Intell. 2012; 26(4): 365-371.

[56] Dale M, O’Keefe S, Sebald A, Stepney S, Trefzer MA. Reservoir computing quality: connectivity and topology. Nat Comput. 2021; 20(2): 205-216.

[57] Manneschi L, Lin AC, Vasilaki E. SpaRCe: improved learning of reservoir computing systems through sparse representations. IEEE Transact Neural Networks Learn Syst. 2023; 34(2): 824-838.

[58] Schrauwen B, Verstraeten D, Van Campenhout J. An overview of reservoir computing: theory, applications and implementations. In:ESANN 2007 Proceedings - 15th European Symposium on Artificial Neural Networks; 2007.

[59] Dominey PF,Ellmore TM, Ventre-Dominey J. Effects of connectivity on narrative temporal processing in structured reservoir computing. In:Proceedings of the International Joint Conference on Neural Networks; 2022.

[60] Shahi S, Fenton FH, Cherry EM. Prediction of chaotic time series using recurrent neural networks and reservoir computing techniques: a comparative study. Mach Learn Appl. 2022; 8:100300.

[61] Bo YC, Wang P, Zhang X. An asynchronously deep reservoir computing for predicting chaotic time series. Appl Soft Comput J. 2020; 95:106530.

[62] Maass W. Liquid state machines:motivation, theory, and applications. In: Computability in Context: Computation and Logic in the Real World; 2011.

[63] Song Q, Feng Z. Effects of connectivity structure of complex echo state network on its prediction performance for nonlinear time series. Neurocomputing. 2010; 73(10-12): 2177-2185.

[64] Polepalli A,Soures N, Kudithipudi D. Reconfigurable digital design of a liquid state machine for spatio-temporal data. In:Proceedings of the 3rd ACM International Conference on Nanoscale Computing and Communication, ACM NANOCOM 2016; 2016.

[65] Lilak S, Woods W, Scharnhorst K, et al. Spoken digit classification by in-materio reservoir computing with neuromorphic atomic switch networks. Front Nanotechnol. 2021; 3.

[66] Sillin HO, Aguilera R, Shieh HH, et al. A theoretical and experimental study of neuromorphic atomic switch networks for reservoir computing. Nanotechnology. 2013; 24(38):384004.

[67] Demis EC, Aguilera R, Scharnhorst K, Aono M, Stieg AZ, Gimzewski JK. Nanoarchitectonic atomic switch networks for unconventional computing. Jpn J Appl Phys. 2016 ; 55(11):1102B2.

[68] Nakajima K. Physical reservoir computing-an introductory perspective. Jpn J Appl Phys. 2020; 59(6):060501.

[69] Mackey MC, Glass L. Oscillation and chaos in physiological control systems. Science. 1977; 197(4300): 287-289.

[70] Doyne Farmer J. Chaotic attractors of an infinite-dimensional dynamical system. Phys D Nonlinear Phenom. 1982; 4(3): 366-393.

[71] Hunt BR, Li T-Y, Kennedy JA, Nusse HE,eds. The Theory of Chaotic Attractors. Springer Science & Business Media; 2013.

[72] Chai Y. In-sensor computing for machine vision. Nature. 2020; 579(7797): 32-33.

[73] Mennel L, Symonowicz J, Wachter S, Polyushkin DK, Molina-Mendoza AJ, Mueller T. Ultrafast machine vision with 2D material neural network image sensors. Nature. 2020; 579(7797): 62-66.

[74] Fayyazi A, Ansari M, Kamal M, Afzali-Kusha A, Pedram M. An ultra low-power memristive neuromorphic circuit for Internet of Things smart sensors. IEEE Internet Things J. 2018; 5(2): 1011-1022.

[75] Chu M, Kim B, Park S, et al. Neuromorphic hardware system for visual pattern recognition with memristor array and CMOS neuron. IEEE Trans Ind Electron. 2015; 62(4): 2410-2419.

[76] Han JK, Yun SY, Lee SW, Yu JM, Choi YK. A review of artificial spiking neuron devices for neural processing and sensing. Adv Funct Mater. 2022; 32(33): 1-17.

[77] Han JK, Geum DM, Lee MW, et al. Bioinspired photoresponsive single transistor neuron for a neuromorphic visual system. Nano Lett. 2020; 20(12): 8781-8788.

[78] Zhou F, Zhou Z, Chen J, et al. Optoelectronic resistive random access memory for neuromorphic vision sensors. Nat Nanotechnol. 2019; 14(8): 776-782.

[79] Yang CM, Chen T, Verma D, et al. Bidirectional all-optical synapses based on a 2D Bi₂O₂Se/graphene hybrid structure for multifunctional optoelectronics. Adv Funct Mater. 2020; 30(30):2001598.

[80] John RA, Acharya J, Zhu C, et al. Optogenetics inspired transition metal dichalcogenide neuristors for in-memory deep recurrent neural networks. Nat Commun. 2020; 11(1):3211.

[81] Zhu QB, Li B, Yang DD, et al. A flexible ultrasensitive optoelectronic sensor array for neuromorphic vision systems. Nat Commun. 2021; 12(1):1798.

[82] Yu J, Yang X, Gao G, et al. Bioinspired mechano-photonic artificial synapse based on graphene/MoS₂ heterostructure. Sci Adv. 2021; 7(12):eabd9117.

[83] Kwon SM, Cho SW, Kim M, Heo JS, Kim Y, Park SK. Environment-adaptable artificial visual perception behaviors using a light-adjustable optoelectronic neuromorphic device array. Adv Mater. 2019; 31(52):1906433.

[84] Feng G, Jiang J, Li Y, Xie D, Tian B, Wan Q. Flexible vertical photogating transistor network with an ultrashort channel for in-sensor visual nociceptor. Adv Funct Mater. 2021; 31(36):2104327.

[85] Kumagai Y, Burton LA, Walsh A, Oba F. Electronic structure and defect physics of tin sulfides: SnS, Sn₂S₃, and Sn S₂. Phys Rev Appl. 2016; 6(1):014009.

[86] Island JO, Blanter SI, Buscema M, Van Der Zant HSJ, Castellanos-Gomez A. Gate controlled photocurrent generation mechanisms in high-gain In₂Se₃ phototransistors. Nano Lett. 2015; 15(12): 7853-7858.

[87] Lee M, Lee W, Choi S, et al. Brain-inspired photonic neuromorphic devices using photodynamic amorphous oxide semiconductors and their persistent photoconductivity. Adv Mater. 2017; 29(28):1700951.

[88] Chen J, Zhou Z, Kim BJ, et al. Optoelectronic graded neurons for bioinspired in-sensor motion perception. Nat Nanotechnol. 2023; 18(8): 882-888.

[89] Jang YH, Han J, Kim J, et al. Graph analysis with multifunctional self-rectifying memristive crossbar array. Adv Mater. 2023; 35(10):2209503.

[90] Chen X, Chen B, Jiang B, et al. Nanowires for UV-vis-IR optoelectronic synaptic devices. Adv Funct Mater. 2022; 2208807: 1-30.

[91] Moin A, Zhou A, Rahimi A, et al. A wearable biosensing system with in-sensor adaptive machine learning for hand gesture recognition. Nat Electron. 2021; 4(1): 54-63.

[92] Ji X, Zhao X, Tan MC, Zhao R. Artificial perception built on memristive system: visual, auditory, and tactile sensations. Adv Intell Syst. 2020; 2(3):1900118.

[93] Röck F, Barsan N, Weimar U. Electronic nose: current status and future trends. Chem Rev. 2008; 108(2): 705-725.

[94] Podrazka M, Báczyńska E, Kundys M, Jeleń PS, Nery EW. Electronic tongue-a tool for all tastes? Biosensors. 2017; 8(1):3.

Funding

National Research Foundation of Korea(2020R1A3B2079882)