Recent Advances in Physical Reservoir Computing: A Review

Abstract

Reservoir computing is a computational framework suited for temporal/sequential data processing. It is derived from several recurrent neural network models, including echo state networks and liquid state machines. A reservoir computing system consists of a reservoir for mapping inputs into a high-dimensional space and a readout for pattern analysis from the high-dimensional states in the reservoir. The reservoir is fixed and only the readout is trained with a simple method such as linear regression and classification. Thus, the major advantage of reservoir computing compared to other recurrent neural networks is fast learning, resulting in low training cost. Another advantage is that the reservoir without adaptive updating is amenable to hardware implementation using a variety of physical systems, substrates, and devices. In fact, such physical reservoir computing has attracted increasing attention in diverse fields of research. The purpose of this review is to provide an overview of recent advances in physical reservoir computing by classifying them according to the type of the reservoir. We discuss the current issues and perspectives related to physical reservoir computing, in order to further expand its practical applications and develop next-generation machine learning systems.

62 pages, 13 figures

References in corpus (32)

• Deep Learning in Neural Networks: An Overview
• Kernel methods in machine learning
• Optoelectronic Reservoir Computing
• Integration of nanoscale memristor synapses in neuromorphic computing architectures
• All-optical Reservoir Computing
• Reinforcement Learning in a large scale photonic Recurrent Neural Network
• Harnessing disordered quantum dynamics for machine learning
• The Fascinating World of Landau-Lifshitz-Gilbert Equation: An Overview
• A magnetic skyrmion as a non-linear resistive element - a potential building block for reservoir computing
• Phase reduction approach to synchronization of nonlinear oscillators
• High performance photonic reservoir computer based on a coherently driven passive cavity
• Distributive Dynamic Spectrum Access through Deep Reinforcement Learning: A Reservoir Computing Based Approach
• Tutorial: Photonic Neural Networks in Delay Systems
• Potential implementation of Reservoir Computing models based on magnetic skyrmions
• Macromagnetic simulation for reservoir computing utilizing spin dynamics in magnetic tunnel junctions
• Reservoir computing with a single time-delay autonomous Boolean node
• Computing with Networks of Nonlinear Mechanical Oscillators
• Memristor - The fictional circuit element
• Online Training of an Opto-Electronic Reservoir Computer Applied to Real-Time Channel Equalisation
• Computational Capabilities of Random Automata Networks for Reservoir Computing
• Brain-inspired photonic signal processor for periodic pattern generation and chaotic system emulation
• Liquid State Machine with Dendritically Enhanced Readout for Low-power, Neuromorphic VLSI Implementations
• Embodiment of Learning in Electro-Optical Signal Processors
• Memristor models for machine learning
• Photonic Delay Systems as Machine Learning Implementations
• Parallel photonic reservoir computing using frequency multiplexing of neurons
• Reservoir Computing using Cellular Automata
• Random pattern and frequency generation using a photonic reservoir computer with output feedback
• Reservoir Computing Hardware with Cellular Automata
• Reservoir Computing Using Non-Uniform Binary Cellular Automata
• A model of reward-modulated motor learning with parallelcortical and basal ganglia pathways
• Chaotic dynamics in a macrospin spin-torque nano-oscillator with delayed feedback

Cited by in corpus (288)

• Deep physical neural networks enabled by a backpropagation algorithm for arbitrary physical systems
• Physical reservoir computing -- An introductory perspective
• An Introduction to Spin Wave Computing
• Large-Scale Optical Reservoir Computing for Spatiotemporal Chaotic Systems Prediction
• Reconfigurable Training and Reservoir Computing in an Artificial Spin-Vortex Ice via Spin-Wave Fingerprinting
• Transformers for Modeling Physical Systems
• The promise of spintronics for unconventional computing
• Opportunities in Quantum Reservoir Computing and Extreme Learning Machines
• Reservoir Computing with Random Skyrmion Textures
• Experimental quantum memristor
• Magnetic Skyrmions for Unconventional Computing
• Embedding and Approximation Theorems for Echo State Networks
• Brownian reservoir computing realized using geometrically confined skyrmions
• Optical Reservoir Computing using multiple light scattering for chaotic systems prediction
• Controlling light propagation in multimode fibers for imaging, spectroscopy and beyond
• Hamiltonian neural networks for solving equations of motion
• Polaritonic neuromorphic computing outperforms linear classifiers
• Dynamical phase transitions in quantum reservoir computing
• Self-organizing memristive nanowire networks with structural plasticity emulate biological neuronal circuits
• Brain-inspired computing with fluidic iontronic nanochannels
• Pattern recognition with neuromorphic computing using magnetic-field induced dynamics of skyrmions
• Temporal Information Processing on Noisy Quantum Computers
• Learning without neurons in physical systems
• Machine learning with neural networks
• Pretrained Transformers as Universal Computation Engines
• Reservoir computing using a spin-wave delay line active ring resonator
• Time Series Quantum Reservoir Computing with Weak and Projective Measurements
• Gaussian states of continuous-variable quantum systems provide universal and versatile reservoir computing
• Natural quantum reservoir computing for temporal information processing
• Pattern recognition in the nucleation kinetics of non-equilibrium self-assembly
• Applications and Techniques for Fast Machine Learning in Science
• Implementing a magnonic time-delay reservoir computer model
• Information Processing Capacity of Spin-Based Quantum Reservoir Computing Systems
• Pattern recognition with a magnon-scattering reservoir
• Exploring the landscapes of "computing": digital, neuromorphic, unconventional -- and beyond
• Biological neurons act as generalization filters in reservoir computing
• Model-free tracking control of complex dynamical trajectories with machine learning
• A Survey on Reservoir Computing and its Interdisciplinary Applications Beyond Traditional Machine Learning
• Combining Machine Learning with Knowledge-Based Modeling for Scalable Forecasting and Subgrid-Scale Closure of Large, Complex, Spatiotemporal Systems
• Reservoir Computing with Diverse Timescales for Prediction of Multiscale Dynamics
• Reconstructing quantum states with quantum reservoir networks
• Spin waves propagating through a stripe magnetic domain structure and their applications to reservoir computing
• Deep Neural Networks using a Single Neuron: Folded-in-Time Architecture using Feedback-Modulated Delay Loops
• Potential and limitations of quantum extreme learning machines
• Coupled nonlinear delay systems as deep convolutional neural networks
• Classification of time-domain waveforms using a speckle-based optical reservoir computer
• High Speed Photonic Neuromorphic Computing Using Recurrent Optical Spectrum Slicing Neural Networks
• Nonlocal stimulation of three-magnon splitting in a magnetic vortex
• Echo State Networks trained by Tikhonov least squares are L2(μ) approximators of ergodic dynamical systems
• Scalable photonic platform for real-time quantum reservoir computing
• Reservoir Computing with Spin Waves in Skyrmion Crystal
• Realising and compressing quantum circuits with quantum reservoir computing
• Chaos on compact manifolds: Differentiable synchronizations beyond the Takens theorem
• Perspective on unconventional computing using magnetic skyrmions
• Deep Photonic Reservoir Computer Based on Frequency Multiplexing with Fully Analog Connection Between Layers
• Breaking Symmetries of the Reservoir Equations in Echo State Networks
• Model-free prediction of multistability using echo state network
• Forecasting the outcome of spintronic experiments with Neural Ordinary Differential Equations
• Multifunctionality in a Reservoir Computer
• Optimizing Memory in Reservoir Computers
• Reservoir Computing Approach to Quantum State Measurement
• Reservoir Computing and Photoelectrochemical Sensors: A Marriage of Convenience
• Transfer learning for nonlinear dynamics and its application to fluid turbulence
• Simulation platform for pattern recognition based on reservoir computing with memristor networks
• Neural Network for Low-Memory IoT Devices and MNIST Image Recognition Using Kernels Based on Logistic Map
• Dissipation as a resource for Quantum Reservoir Computing
• Learning Hamiltonian dynamics by reservoir computer
• Robust Forecasting using Predictive Generalized Synchronization in Reservoir Computing
• Fundamental aspects of noise in analog-hardware neural networks
• Feedback-driven quantum reservoir computing for time-series analysis
• Insight into Delay Based Reservoir Computing via Eigenvalue Analysis
• Neural Schrödinger Equation:Physical Law as Neural Network
• Optimized ensemble deep learning framework for scalable forecasting of dynamics containing extreme events
• Neuromorphic Overparameterisation and Few-Shot Learning in Multilayer Physical Neural Networks
• Learning strange attractors with reservoir systems
• Reservoir Computing Benchmarks: a tutorial review and critique
• Creating and concentrating quantum resource states in noisy environments using a quantum neural network
• Unveiling the role of plasticity rules in reservoir computing
• Audio Classification with Skyrmion Reservoirs
• RC-Struct: A Structure-based Neural Network Approach for MIMO-OFDM Detection
• Reservoir Computing with Noise
• SpaRCe: Improved Learning of Reservoir Computing Systems through Sparse Representations
• Exploring quantum mechanical advantage for reservoir computing
• Learn to Synchronize, Synchronize to Learn
• Quantum reservoir computing in finite dimensions
• Predicting Critical Transitions in Multiscale Dynamical Systems Using Reservoir Computing
• Model-Size Reduction for Reservoir Computing by Concatenating Internal States Through Time
• Physical deep learning based on optimal control of dynamical systems
• Overcoming the Coherence Time Barrier in Quantum Machine Learning on Temporal Data
• Energy time series forecasting-Analytical and empirical assessment of conventional and machine learning models
• Moire synaptic transistor for homogeneous-architecture reservoir computing
• How quantum computing can enhance biomarker discovery
• Noisy Quantum Kernel Machines
• In-materio neuromimetic devices: Dynamics, information processing and pattern recognition
• Reducing network size and improving prediction stability of reservoir computing
• A numerical exploration of signal detector arrangement in a spin-wave reservoir computing device
• Biomembrane-based Memcapacitive Reservoir Computing System for Energy Efficient Temporal Data Processing
• Reservoir Computing for Macroeconomic Forecasting with Mixed Frequency Data
• Handwritten Digit Recognition by Spin Waves in a Skyrmion Reservoir
• Machine Learning Potential of a Single Pendulum
• Time Shifts to Reduce the Size of Reservoir Computers
• Chaos may enhance expressivity in cerebellar granular layer
• Reservoirs learn to learn
• A Method for Medical Data Analysis Using the LogNNet for Clinical Decision Support Systems and Edge Computing in Healthcare
• Dynamical robustness of network of oscillators
• Ultrafast single-channel machine vision based on neuro-inspired photonic computing
• Tackling Sampling Noise in Physical Systems for Machine Learning Applications: Fundamental Limits and Eigentasks
• Machine Learning Link Inference of Noisy Delay-coupled Networks with Opto-Electronic Experimental Tests
• Implementing an insect brain computational circuit using III-V nanowire components in a single shared waveguide optical network
• Optimizing quantum noise-induced reservoir computing for nonlinear and chaotic time series prediction
• An Efficient and Accurate Memristive Memory for Array-based Spiking Neural Networks
• Photons guided by axons may enable backpropagation-based learning in the brain
• Reservoir Computing based on Quenched Chaos
• Hybridizing Traditional and Next-Generation Reservoir Computing to Accurately and Efficiently Forecast Dynamical Systems
• Criticality in Reservoir Computer of Coupled Phase Oscillators
• Harnessing Quantum Extreme Learning Machines for image classification
• Perceptron Theory Can Predict the Accuracy of Neural Networks
• Superpolynomial Quantum Enhancement in Polaritonic Neuromorphic Computing
• Transfer learning of chaotic systems
• Echo State Network for two-dimensional turbulent moist Rayleigh-Bénard convection
• Time-Delayed Reservoir Computing Based on Dual-Waveband Quantum-Dot Spin-Polarized Vertical Cavity Surface-Emitting Laser
• Delay-induced self-oscillation excitation in the FitzHugh-Nagumo model: regular and chaotic dynamics
• Extending echo state property for quantum reservoir computing
• Uncovering multifunctional mechano-intelligence in and through phononic metastructures harnessing physical reservoir computing
• Benchmarking the role of particle statistics in Quantum Reservoir Computing
• Effective models and predictability of chaotic multiscale systems via machine learning
• Towards synthetic neural networks: Can artificial electrochemical neurons be coupled with artificial memristive synapses?
• Protein Structured Reservoir computing for Spike-based Pattern Recognition
• Reservoir Computing with Generalized Readout based on Generalized Synchronization
• Skyrmion based energy efficient straintronic physical reservoir computing
• State estimation with quantum extreme learning machines beyond the scrambling time
• Spatial Analysis of Physical Reservoir Computers
• Passive frustrated nanomagnet reservoir computing
• Training Coupled Phase Oscillators as a Neuromorphic Platform using Equilibrium Propagation
• Optical Neural Network Based on Synthetic Nonlinear Photonic Lattices
• Deep Photonic Reservoir Computer for Speech Recognition
• Emergent self-adaptation in an integrated photonic neural network for backpropagation-free learning
• Short-wavelength Reverberant Wave Systems for Physical Realization of Reservoir Computing
• Experimental demonstration of bandwidth enhancement in photonic time delay reservoir computing
• Retrieving past quantum features with deep hybrid classical-quantum reservoir computing
• Creating New Chaotic Signals with Reservoir Computers
• Squeezing as a resource for time series processing in quantum reservoir computing
• Nonlinear Inference Capacity of Fiber-Optical Extreme Learning Machines
• Learning unidirectional coupling using echo-state network
• Neural Echo State Network using oscillations of gas bubbles in water
• Photonic kernel machine learning for ultrafast spectral analysis
• P-CRITICAL: A Reservoir Autoregulation Plasticity Rule for Neuromorphic Hardware
• Practical Few-Atom Quantum Reservoir Computing
• Reservoir Computing meets Recurrent Kernels and Structured Transforms
• Reservoir Computing with Superconducting Electronics
• Quantum reservoir probing of quantum phase transitions
• Current-Induced Dynamics of Chiral Magnetic Structures: Creation, Motion, and Applications
• Artificial Spin Ice: A Tutorial on Design and Control of Geometry, Microstate, Magnon Dynamics & Neuromorphic Computing
• Beating the classical phase precision limit using a quantum neuromorphic platform
• Stochastic Reservoir Computers
• Role of assortativity in predicting burst synchronization using echo state network
• Complex-valued reservoir computing for aspect classification and slope-angle estimation with low computational cost and high resolution in interferometric synthetic aperture radar
• Improving the Performance of Echo State Networks Through State Feedback
• Mode-resolved micromagnetics study of parametric spin wave excitation in thin-film disks
• The Computational Capacity of LRC, Memristive and Hybrid Reservoirs
• Stochastic approach for assessing the predictability of chaotic time series using reservoir computing
• Reservoir Computing using High Order Synchronization of Coupled Oscillators
• Dynamical analysis of a parameter-aware reservoir computer
• Physical Neural Networks with Self-Learning Capabilities
• Artificial Nanophotonic Neuron with Internal Memory for Biologically Inspired and Reservoir Network Computing
• Dynamical Systems as Temporal Feature Spaces
• Neuromorphic computing with a single qudit
• Dimensions of Timescales in Neuromorphic Computing Systems
• Choose your tools carefully: A Comparative Evaluation of Deterministic vs. Stochastic and Binary vs. Analog Neuron models for Implementing Emerging Computing Paradigms
• Recurrent Spectral Network (RSN): shaping the basin of attraction of a discrete map to reach automated classification
• Time-Varying Convex Optimization: A Contraction and Equilibrium Tracking Approach
• Optofluidic memory and self-induced nonlinear optical phase change for reservoir computing in silicon photonics
• Oscillations enhance time-series prediction in reservoir computing with feedback
• Ciliary flocking and emergent instabilities enable collective agility in a non-neuromuscular animal
• Quantum reservoir computing on random regular graphs
• Time-shift selection for reservoir computing using a rank-revealing QR algorithm
• Recognition of handwritten MNIST digits on low-memory 2 Kb RAM Arduino board using LogNNet reservoir neural network
• Non-spectroscopic sensing enabled by electro-optical reservoir computer
• Reservoir Computing with Planar Nanomagnet Arrays
• Initializing LSTM internal states via manifold learning
• Thermally-robust spatiotemporal parallel reservoir computing by frequency filtering in frustrated magnets
• Transport in reservoir computing
• Quantum-tunnelling deep neural network for optical illusion recognition
• Quantum Next-Generation Reservoir Computing and Its Quantum Optical Implementation
• Atom-by-atom construction of attractors in a tunable finite size spin array
• Higher Order and Long-Range Synchronization Effects for Classification and Computing in Oscillator-Based Spiking Neural Networks
• Chaotic magnetization dynamics in magnetic Duffing oscillator
• Conduction modulation of solution-processed two-dimensional materials
• Spintronic Physical Reservoir for Autonomous Prediction and Long-Term Household Energy Load Forecasting
• Understanding Recurrent Neural Networks Using Nonequilibrium Response Theory
• Quantum reservoir probing: an inverse paradigm of quantum reservoir computing for exploring quantum many-body physics
• Universal Approximation Theorem and error bounds for quantum neural networks and quantum reservoirs
• Blending Optimal Control and Biologically Plausible Learning for Noise-Robust Physical Neural Networks
• Magneto-Ionic Physical Reservoir Computing
• Physical Implementation of a Tunable Memristor-based Chua's Circuit
• Optimization of a Hydrodynamic Computational Reservoir through Evolution
• Predicting the motion of a high-Q pendulum subject to seismic perturbations using machine learning
• Low frequency noise in nanoparticle-molecule networks and implications for in-materio reservoir computing
• Universality and approximation bounds for echo state networks with random weights
• Toward the Fully Physics-Informed Echo State Network -- an ODE Approximator Based on Recurrent Artificial Neurons
• Hardware-Friendly Delayed-Feedback Reservoir for Multivariate Time-Series Classification
• ReLiCADA -- Reservoir Computing using Linear Cellular Automata Design Algorithm
• Practical Trainable Temporal Postprocessor for Multistate Quantum Measurement
• Exploring Non-Steady-State Charge Transport Dynamics in Information Processing: Insights from Reservoir Computing
• Unsupervised Reservoir Computing for Solving Ordinary Differential Equations
• Photonic neural field on a silicon chip: large-scale, high-speed neuro-inspired computing and sensing
• Learning Beyond Experience: Generalizing to Unseen State Space with Reservoir Computing
• Universality of reservoir systems with recurrent neural networks
• Estimating many properties of a quantum state via quantum reservoir processing
• Excitable mechanics embodied in a walking cilium
• Model-free Forecasting of Rogue Waves using Reservoir Computing
• Forecasting Using Reservoir Computing: The Role of Generalized Synchronization
• Physical Reservoir Computing with Origami and its Application to Robotic Crawling
• Improving the prediction of spatio-temporal chaos by combining parallel reservoir computing with dimensionality reduction
• Modeling Defect-Level Switching for Highly-Nonlinear and Hysteretic Electronic Devices
• Restrictions on Physical Stochastic Reservoir Computers
• AdvectiveNet: An Eulerian-Lagrangian Fluidic reservoir for Point Cloud Processing
• The design, verification, and applications of Hotspice: a Monte Carlo simulator for artificial spin ice
• Task learning through stimulation-induced plasticity in neural networks
• High-Accuracy Temporal Prediction via Experimental Quantum Reservoir Computing in Correlated Spins
• SuperNet -- An efficient method of neural networks ensembling
• An introduction to reservoir computing
• Echoes of the Past: A Unified Perspective on Fading memory and Echo States
• Enhanced Predictive Capability for Chaotic Dynamics by Modified Quantum Reservoir Computing
• Fast gradient-free activation maximization for neurons in spiking neural networks
• Unsupervised Detection of Topological Phase Transitions with a Quantum Reservoir
• Analysis on reservoir activation with the nonlinearity harnessed from solution-processed molybdenum disulfide
• Generating Quantum Reservoir State Representations with Random Matrices
• Next-Generation Reservoir Computing for Dynamical Inference
• Echo state property and memory capacity of artificial spin ice
• Machine learning via relativity-inspired quantum dynamics
• Information processing via human soft tissue
• Crossover between kite growth and vibrational bridging in pillar-assisted controlled formation of carbon nanotube networks
• Learning Augmentation Distributions using Transformed Risk Minimization
• Neural information processing and time-series prediction with only two dynamical memristors
• Perpendicular-anisotropy artificial spin ice with spontaneous ordering: a platform for neuromorphic computing with flexible timescales
• An experimental demonstration of neuromorphic sensing of chemical species using electro-optical reservoir computing
• Harnessing Quantum Dynamics for Robust and Scalable Quantum Extreme Learning Machines
• Phase transitions from linear to nonlinear information processing in neural networks
• Edge of Many-Body Quantum Chaos in Quantum Reservoir Computing
• A General Molecular-Scale Dynamic Memristor Model Based on Non-equilibrium Charge Transport Kinetics and Its Information Processing Capability in Reservoir Computing
• Quantum Transport Reservoir Computing
• Metrics for spin-based computing
• Prediction performance of random reservoirs with different topology for nonlinear dynamical systems with different number of degrees of freedom
• Cellular Memristive-Output Reservoir (CMOR)
• Data-driven stability analysis in a multi-element supercritical Liquid Oxygen-methane combustor
• Transfer between long-term and short-term memory using Conceptors
• Excitatory/Inhibitory Balance Emerges as a Key Factor for RBN Performance, Overriding Attractor Dynamics
• Photonic reservoir computing based on nonlinear wave dynamics at a microscale
• Rebooting Neuromorphic Hardware Design -- A Complexity Engineering Approach
• Spoken digit classification using a spin-wave delay-line active-ring reservoir computing
• On Artificial Life and Emergent Computation in Physical Substrates
• Frequency-tunable nano-oscillator based on Ovonic Threshold Switch (OTS)
• Fast Parameter Optimization of Delayed Feedback Reservoir with Backpropagation and Gradient Descent
• The Role of Entanglement in Quantum Reservoir Computing with Coupled Kerr Nonlinear Oscillators
• Online Training and Inference System on Edge FPGA Using Delayed Feedback Reservoir
• Building Reservoir Computing Hardware Using Low Energy-Barrier Magnetics
• Residual Reservoir Memory Networks
• Reservoir Transformers
• Information, communication and music: Recognition of musical dissonance and consonance in a simple reservoir computing system
• Cellular automata can classify data by inducing trajectory phase coexistence
• Dynamic Reservoir Computing with Physical Neuromorphic Networks
• Shaping dynamical neural computations using spatiotemporal constraints
• Regulating protocols of globally coupled continuum systems: Effects on dynamics and thermodynamics
• (Thesis) Reservoir Computing With Dynamical Systems
• 56 GBaud PAM-4 100 km Transmission System with Photonic Processing Schemes
• Spatially distributed computation in cortical circuits
• 2022 Roadmap on Neuromorphic Computing and Engineering
• A Small-Gain Theorem for Discrete-Time Convergent Systems and Its Applications
• Nonlinear Autoregression with Convergent Dynamics on Novel Computational Platforms
• Predicting unobserved climate time series data at distant areas via spatial correlation using reservoir computing
• Bridging the Band Gap: What Device Physicists Need to Know About Machine Learning
• Physical reservoir computing using finitely-sampled quantum systems
• Orthogonal Transforms in Neural Networks Amount to Effective Regularization
• Low-Dimensional Manifolds Support Multiplexed Integrations in Recurrent Neural Networks
• Computation harvesting in road traffic dynamics
• A silicon photonics feed-forward neural network for nonlinear distortion mitigation in an optical link
• A Deep Reinforcement Learning Trader without Offline Training
• Quantum reservoir computing in Jaynes-Cummings models: Nonlinear memory and time-series prediction
• Topological phase transitions by time-dependent electromagnetic fields in frustrated magnets: Role of dynamical and static magnetic fields
• Focused Surface Acoustic Wave induced nano-oscillator based reservoir computing
• Learning-Performance Evaluation of a Physical Reservoir Based on a Vortex Spin-Torque Oscillator with a Modified Free Layer
• Medical Imaging Classification with Cold-Atom Reservoir Computing using Auto-Encoders and Surrogate-Driven Training
• Memory Uncertainty Relation and Harmonic Memory in Random Recurrent Networks
• A magnonic-optoelectronic reservoir for physical reservoir computing
• Reservoir computing by thin film embedded with magnetic impurities
• Experimental Investigation of Time Series Classification using a Self-Pulsing Microring Resonator Network
• Using multidimensional speckle dynamics for high-speed, large-scale, parallel photonic computing