Reservoir Computing Networks Research Articles

Enhanced Ferroelectricity of Hf‐Based Memcapacitors by Adopting Ti Insert‐Layer and C–V Measurement for Constructing Energy‐Efficient Reservoir Computing Network

Enhanced Ferroelectricity of Hf‐Based Memcapacitors by Adopting Ti Insert‐Layer and C–V Measurement for Constructing Energy‐Efficient Reservoir Computing Network

Exploration of a brain-inspired photon reservoir computing network based on quantum-dot spin-VCSELs.

Based on small-world network theory, we have developed a brain-inspired photonic reservoir computing (RC) network system utilizing quantum dot spin-vertical-cavity surface-emitting lasers (QD spin-VCSELs) and formulated a comprehensive theoretical model for it. This innovative network system comprises input layers, a reservoir network layer, and output layers. The reservoir network layer features four distinct reservoir modules that are asymmetrically coupled. Each module is represented by a QD spin-VCSEL, characterized by optical feedback and optical injection. Within these modules, four chaotic polarization components, emitted from both the ground and excited states of the QD Spin-VCSEL, form four distinct reservoirs through a process of asymmetric coupling. Moreover, these components, when emitted by the ground and excited states of a driving QD spin-VCSEL within a specific parameter space, act as targets for prediction. Delving further, we investigated the correlation between various system parameters, such as the sampling period, the interval between virtual nodes, the strengths of optical injection and feedback, frequency detuning, and the predictive accuracy of each module's four photonic RCs concerning the four designated predictive targets. We also examined how these parameters influence the memory storage capabilities of the four photonics RCs within each module. Our findings indicate that when a module receives coupling injections from more than two other modules, and an RC within this module is also subject to coupling injections from over two other RCs, the system displays reduced predictive errors and enhanced memory storage capacities when the system parameters are fixed. Namely, the superior performance of the reservoir module in predictive accuracy and memory capacities follows from its complex interaction with multiple light injections and coupling injections, with its three various PCs benefiting from three, two, and one coupling injections respectively. Conversely, variations in optical injection and feedback strength, as well as frequency detuning, introduce only marginal fluctuations in the predictive errors across the four photonics RCs within each module and exert minimal impact on the memory storage capacity of individual photonics RCs within the modules. Our investigated results contribute to the development of photonic reservoir computing towards fast response biological neural networks.

Comparison of machine learning systems trained to detect Alfvén eigenmodes using the CO2 interferometer on DIII-D

A Machine-Learning (ML) based detection scheme that automatically detects Alfvén Eigenmodes (AE) in a labelled DIII-D database is presented here. Controlling AEs is important for the success of planned burning plasma devices such as ITER, since resonant fast ions can drive AEs unstable and degrade the performance of the plasma or damage the first walls of the machine vessel. Artificial Intelligence could be useful for real-time detection and control of AEs in steady-state plasma scenarios by implementing ML-based models into control algorithms that drive actuators for mitigation of AE impacts. Thus, the objective is to compare differences in performance between using two different recurrent neural network systems (Reservoir Computing Network and Long Short Term Memory Network) and two different representations of the CO2 phase data (simple and crosspower spectrograms). All CO2 interferometer chords are used to train both models, but only one is processed during each training step. The results from the model and data comparison show higher performance for the RCN model (True Positive Rate = 90% and False Positive Rate = 14%), and that using simple magnitude spectrograms is sufficient to detect AEs. Also, the vertical CO2 interferometer chord passing near the center is better for ML-based detection of AEs.

Adjustable short-term memory of SiO x :Ag-based memristor for reservoir computing

Temporal information processing is critical for a wide spectrum of applications, such as finance, biomedicine, and engineering. Reservoir computing (RC) can efficiently process temporal information with low training costs. Various memristors have been explored to demonstrate RC systems leveraging the short-term memory and nonlinear dynamic behaviours. However, the short-term memory is fixed after the device fabrication, limiting the applications to diverse temporal analysis tasks. In this work, we propose the approaches to modulating the short-term memory of Pt/SiO x :Ag/Pt memristor for the performance improvement of the RC systems. By controlling the read voltage, pulse amplitude and pulse width applied to the devices, the obtainable range of the characteristic time reaches three orders of magnitude from microseconds to around milliseconds. Based on the fabricated memristor, the classification of 4-bit pulse streams is demonstrated. Memristor-based RC systems with adjustable short-term memory are constructed for time-series prediction and pattern recognition tasks with different requirements for the characteristic times. The simulation results show that low normalized root mean square error of 0.003 (0.27) in Hénon map (Mackey–Glass time series) and excellent classification accuracy of 99.6% (91.7%) in spoken-digit recognition (MNIST image recognition) are achieved, which outperforms most memristor-based RC systems recently reported. Furthermore, the RC networks with diverse short-term memories are constructed to address more complicated tasks with low prediction errors. This work proves the high controllability of memristor-based RC systems to handle multiple temporal processing tasks.

An energy efficient reservoir computing system based on HZO memcapacitive devices

Memcapacitor devices based on ferroelectric material have attracted attention recently in application of neuromorphic computing due to lower static power relative to memristors. They have been used for establishing fully connected neural networks but not yet for recurrent neural networks (RNNs), which owns the advantage in temporal signal processing. As an improved network architecture for RNNs, reservoir computing (RC) is easier to train and energy efficient. In this work, an HZO-based ferroelectric memcapacitor is used as the reservoir layer to recognize handwritten digits. A recognition accuracy of 90.3% is achieved. Meanwhile, a task of predicting Mackey–Glass time series is built to demonstrate the advantage of reservoir networks in processing time-series signals. A normalized root mean square error of 0.13 was obtained, indicating that this system can predict the Mackey–Glass chaotic system well. In addition, the energy consumption in the input signal and recognition task is significantly lowered compared with a memristor-based network. Our work provides an energy efficient way to build up the RC network.

Interpretable Design of Reservoir Computing Networks Using Realization Theory.

The reservoir computing networks (RCNs) have been successfully employed as a tool in learning and complex decision-making tasks. Despite their efficiency and low training cost, practical applications of RCNs rely heavily on empirical design. In this article, we develop an algorithm to design RCNs using the realization theory of linear dynamical systems. In particular, we introduce the notion of α -stable realization and provide an efficient approach to prune the size of a linear RCN without deteriorating the training accuracy. Furthermore, we derive a necessary and sufficient condition on the irreducibility of the number of hidden nodes in linear RCNs based on the concepts of controllability and observability from systems theory. Leveraging the linear RCN design, we provide a tractable procedure to realize RCNs with nonlinear activation functions. We present numerical experiments on forecasting time-delay systems and chaotic systems to validate the proposed RCN design methods and demonstrate their efficacy.

Fully Flash-Based Reservoir Computing Network With Low Power and Rich States

Fully Flash-Based Reservoir Computing Network With Low Power and Rich States

Deep reservoir computing based on self-rectifying memristor synapse for time series prediction

Herein, a self-rectifying resistive switching memristor synapse with a Ta/NbOx/Pt structure was demonstrated for deep reservoir computing (RC). The memristor demonstrated stable nonlinear analog switching characteristics, with a rectification ratio of up to 1.6 × 105, good endurance, and high uniformity. Additionally, the memristor exhibited typical short-term plasticity and dynamic synaptic characteristics. Based on these characteristics, a deep memristor RC system was proposed for time series prediction. The system achieved a low normalized root mean square error (NRMSE) of 0.04 in the time series prediction of the Henon map. Even at 90 °C, deep RC retains good predictive power with an NRMSE of only 0.07. This work provides guidance for efficient deep memristive RC networks to handle more complex future temporal tasks.

1D and 2D Chaotic Time Series Prediction Using Hierarchical Reservoir Computing System

Reservoir Computing (RC) is a type of machine learning inspired by neural processes, which excels at handling complex and time-dependent data while maintaining low training costs. RC systems generate diverse reservoir states by extracting features from raw input and projecting them into a high-dimensional space. One key advantage of RC networks is that only the readout layer needs training, reducing overall training expenses. Memristors have gained popularity due to their similarities to biological synapses and compatibility with hardware implementation using various devices and systems. Chaotic events, which are highly sensitive to initial conditions, undergo drastic changes with minor adjustments. Cascade chaotic maps, in particular, possess greater chaotic properties, making them difficult to predict with memoryless devices. This study aims to predict 1D and 2D cascade chaotic time series using a memristor-based hierarchical RC system.

Performance Assessment of Joint Optical-Digital Nonlinearity Mitigation Schemes in Long-Haul Systems

The combined approach of optical and digital nonlinearity mitigation techniques have been shown to have an edge over the individual schemes, tackling their drawbacks and improving compensation capabilities. We demonstrated fiber-nonlinearity mitigation of 32 GBd single-polarization 16QAM signals transmitted over an 800-km dispersion-managed link using optical phase conjugation (OPC) and digital-domain neural networks (NN). Our implemented NN comprised a recurrent neural network (RNN), and a reservoir computing network (RCN). To further compensate for the penalty introduced by the design of OPC, we employed NN schemes in the digital signal processing and applied it to the signal after transmission in the OPC-assisted link. We present the results for optimizing the NN models for important hyper-parameters like the number of hidden layer neurons and the input vector size. The proposed joint approach achieves Q <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> -factor improvements up to 1.8 dB while surpassing the improvements of the individual schemes. Our experiments indicate that the joint approach has the potential to reduce the overall complexity of NN architectures in terms of the size of the hidden layer and input vector.

Fully Ferroelectric-FETs Reservoir Computing Network for Temporal and Random Signal Processing

Reservoir computing (RC), a derivation of recurrent neural networks (RNNs), is an energy-efficient computational framework suitable for temporal signal processing. Owing to the short-term and long-term memory capability, the ferroelectric field-effect transistor (FeFET) is regarded as a promising hardware component for implementing RC networks. This article aims to optimize the fully FeFETs RC network by evaluating the recognition accuracy in various classification tasks, which includes the operating voltage sequence, device numbers as well as connection methods. The physical random telegraph noise (RTN), working as an ideal temporal and random signal, is investigated and extended by using the optimized fully FeFET RC network, resulting in a rapid time constant extraction method. Our findings may provide the broad potential for hardware security and cyber security based on the fully FeFET RC network.

A Compact Fully Ferroelectric-FETs Reservoir Computing Network With Sub-100 ns Operating Speed

Reservoir computing (RC) is a low-cost and temporary-signal friendly computational framework, whose hardware implementation is hindered by integrating huge amounts and various kinds of devices. Benefitted from the logic in memory (LIM) capability, the process compatible Hf <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0.5</sub> Zr <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0.5</sub> O <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> (HZO)-based ferroelectric field-effect-transistor (FeFET) is a promising candidate for implementing artificial networks. In this letter, a fully FeFETs RC network is proposed. Multiple functions can be achieved in a single device owing to its intrinsic characteristics, and the richness of virtual nodes is largely enhanced through full-connection structures. Impressively, only 44 FeFETs are required to construct a compact RC network with 100 ns operating speed and high accuracy in classification tasks. This paves the way to develop the high energy-efficiency FeFET RC networks.

Time series change detection using reservoir computing networks for remote sensing data

In this study, a multivariate time series forecasting model applicable to high-dimensional short-term forecasting is proposed. The Reservoir Computing Network (RCN) state reduction method is investigated to address the dimensionality arising from high-dimensional data. To enable the states to express very distant dependencies in time so that the RCN satisfies the echo state property, a bidirectional RCN model is investigated to capture the dependencies of the data forward and backward in time. To solve the short-term data prediction, the spatiotemporal transformation theory is combined with the RCN to exploit the conjugate property between delayed and nondelayed embeddings of dynamical systems, and the correlation between variables in high-dimensional systems is used to compensate for the shortcomings of short-term series, design the method of the RCN training, and complete the prediction of future states. In the experiments, four existing models based on two different data sets were used for comparison, and the results showed that our method has better prediction performance.

PyRCN: A toolbox for exploration and application of Reservoir Computing Networks

Reservoir Computing Networks (RCNs) belong to a group of machine learning techniques that project the input space non-linearly into a high-dimensional feature space, where the underlying task can be solved linearly. Popular variants of RCNs are capable of solving complex tasks equivalently to widely used deep neural networks, but with a substantially simpler training paradigm based on linear regression. In this paper, we show how to uniformly describe RCNs with small and clearly defined building blocks, and we introduce the Python toolbox PyRCN (Python Reservoir Computing Networks) for optimizing, training and analyzing RCNs on arbitrarily large datasets. The tool is based on widely-used scientific packages and complies with the scikit-learn interface specification. It provides a platform for educational and exploratory analyses of RCNs, as well as a framework to apply RCNs on complex tasks including sequence processing. With a small number of building blocks, the framework allows the implementation of numerous different RCN architectures. We provide code examples on how to set up RCNs for time series prediction and for sequence classification tasks. PyRCN is around ten times faster than reference toolboxes on a benchmark task while requiring substantially less boilerplate code.

Joint feature enhancement mapping and reservoir computing for improving fault diagnosis performance

Complicated industrial robot structure and harsh working conditions may cause signal features collected in the condition monitoring process to be seriously disturbed. In this paper, a joint feature enhancement mapping and reservoir computing (FEM-RC) method is presented to handle the industrial robot fault diagnosis problem. Firstly, a feature enhancement mapping (FEM) method is proposed to achieve intraclass distance minimization and interclass distance equalization to obtain an enhanced feature matrix. Then, the first reservoir computing (RC) network is adopted to map the original feature matrix to the feature enhancement matrix, and the second RC network is for fault type classification. The results of the experiment carried out on a six-axial industrial robot demonstrate that compared with other peer models, the present FEM-RC has better fault diagnosis performance and robustness.

Quantum Neuromorphic Computing with Reservoir Computing Networks

AbstractQuantum reservoir networks combine the intelligence of neural networks with the potential of quantum computing in a single platform. This platform operates on the architecture of reservoir computing, which can function even with random connections between neural nodes. This is a major advantage for hardware implementation. Herein is described how reservoir computing is brought into the quantum domain to perform various tasks, including characterization of quantum states, quantum estimation, quantum state preparation, and quantum computing. It shows quantum enhancement in classical data processing, and creates the opportunity for quantum information processing within the robust paradigm of neural networks. It is friendly for implementation in a wide range of physical systems, including quantum dots, superconductors, trapped ions, cold atoms, and exciton‐polaritons.

Protein Structured Reservoir Computing for Spike-Based Pattern Recognition

Nowadays we witness a miniaturisation trend in the semiconductor industry backed up by groundbreaking discoveries and designs in nanoscale characterisation and fabrication. To facilitate the trend and produce ever smaller, faster and cheaper computing devices, the size of nanoelectronic devices is now reaching the scale of atoms or molecules - a technical goal undoubtedly demanding for novel devices. Following the trend, we explore an unconventional route of implementing a reservoir computing on a single protein molecule and introduce neuromorphic connectivity with a small-world networking property. We have chosen Izhikevich spiking neurons as elementary processors, corresponding to the atoms of verotoxin protein, and its molecule as a 'hardware' architecture of the communication networks connecting the processors. We apply on a single readout layer various training methods in a supervised fashion to investigate whether the molecular structured Reservoir Computing (RC) system is capable to deal with machine learning benchmarks. We start with the Remote Supervised Method, based on Spike-Timing-Dependent-Plasticity, and carry on with linear regression and scaled conjugate gradient back-propagation training methods. The RC network is evaluated as a proof-of-concept on the handwritten digit images from the MNIST dataset and demonstrates acceptable classification accuracy in comparison with other similar approaches.

Invertible generalized synchronization: A putative mechanism for implicit learning in neural systems.

Regardless of the marked differences between biological and artificial neural systems, one fundamental similarity is that they are essentially dynamical systems that can learn to imitate other dynamical systems whose governing equations are unknown. The brain is able to learn the dynamic nature of the physical world via experience; analogously, artificial neural systems such as reservoir computing networks (RCNs) can learn the long-term behavior of complex dynamical systems from data. Recent work has shown that the mechanism of such learning in RCNs is invertible generalized synchronization (IGS). Yet, whether IGS is also the mechanism of learning in biological systems remains unclear. To shed light on this question, we draw inspiration from features of the human brain to propose a general and biologically feasible learning framework that utilizes IGS. To evaluate the framework's relevance, we construct several distinct neural network models as instantiations of the proposed framework. Regardless of their particularities, these neural network models can consistently learn to imitate other dynamical processes with a biologically feasible adaptation rule that modulates the strength of synapses. Further, we observe and theoretically explain the spontaneous emergence of four distinct phenomena reminiscent of cognitive functions: (i) learning multiple dynamics; (ii) switching among the imitations of multiple dynamical systems, either spontaneously or driven by external cues; (iii) filling-in missing variables from incomplete observations; and (iv) deciphering superimposed input from different dynamical systems. Collectively, our findings support the notion that biological neural networks can learn the dynamic nature of their environment through the mechanism of IGS.

Possible Mechanism of Internal Visual Perception: Context-dependent Processing by Predictive Coding and Reservoir Computing Network

The predictive coding is a widely accepted hypothesis on how our internal visual perceptions are generated. Dynamical predictive coding with reservoir computing (PCRC) models have been proposed, but how they work remains to be clarified. Therefore, we first construct a simple PCRC network and analyze the nonlinear dynamics underlying it. Since the influence of contexts is another important factor on the visual perception, we also construct PCRC networks for the context-dependent task, and observe their attractor-landscapes on each context.

Possible Mechanism of Internal Visual Perception: Context-dependent Processing by Predictive Coding and Reservoir Computing Network

Possible Mechanism of Internal Visual Perception: Context-dependent Processing by Predictive Coding and Reservoir Computing Network