Top Research Papers on Neuromorphic Computing

To understand the theoretical underpinnings of the neuromorphic approach and predict the likelihood of its implementation within the next decade, it is vital to understand the rise of high performance and Quantum computing as promising alternatives.

This work introduces SONIC, a software-defined hardware design methodology to make neuromorphic computing accessible to the general computing community, and evaluates SONIC using three spiking datasets.

This review summarizes CMOS-compatible neuromorphic devices and discusses their emulation of synaptic and neuronal functions as well as their applications in neuromorphic perception and computing.

A compact and versatile neuromorphic computing core that integrates 1K neurons and 1M synapses through 588 LUTs is built through the use of neuron multiplexing technology and weight clustering algorithm.

The last few decades have witnessed the rapid development of electronic computers relying on von Neumann architecture. However, due to the spatial separation of the memory unit from the computing processor, continuous data movements between them result in intensive time and energy consumptions, which unfortunately hinder the further development of modern computers. Inspired by biological brain, the in situ computing of memristor architectures, which has long been considered to hold unprecedented potential to solve the von Neumann bottleneck, provides an alternative network paradigm for the nex...

The state‐of‐the‐art spintronic technologies, such as the magnetic tunnel junction, spin–orbit torque, domain wall propagation, magnetic skyrmions, and antiferromagnet, are highlighted and how they can used for artificial neurons and synapses in different artificial neural networks are discussed.

The recent advances of ONNs based on different approaches to photonic multiplexing are reviewed, and the outlook on key technologies needed to further advance these photonicMultiplexing/hybrid-multiplexing techniques of Onns are presented.

The aim of this roadmap is to present a snapshot of the present state of neuromorph technology and provide an opinion on the challenges and opportunities that the future holds in the major areas of neuromorphic technology, namely materials, devices, neuromorphic circuits, neuromorph algorithms, applications, and ethics.

The journal ``Neuromorphic Computing and Engineering'' (NCE) has been launched to support this new community in this effort and provide a forum and repository for presenting and discussing its latest advances.

This work proposes utilizing the F1Tenth platform as an evaluation task for neuromorphic computing, and presents a workflow with neuromorphic hardware, software, and training that can be used to develop a spiking neural network for neuromorph hardware deployment to perform autonomous racing.

This work describes 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.

Neuromorphic computation is based on memristors, which function equivalently to neurons in brain structures. These memristors can be made more efficient and tailored to neuromorphic devices by using ferroelastic domain boundaries as fast diffusion paths for ionic conduction, such as of oxygen, sodium, or lithium. In this paper, we show that the local memristor generates a second, unexpected feature, namely, weak magnetic fields that emerge from moving ferroelastic needle domains and vortices. The vortices appear near ferroelastic “junctions” that are common when the external stimulus is a comb...

This work outlines the principle of Neuromorphic imaging, which makes use of an event sensor that responds to changes in pixel intensities, and discusses the computational algorithms to process such event data for a variety of applications.

In this paper, we highlight how computational properties of biological dendrites can be leveraged for neuromorphic applications. Specifically, we demonstrate analog silicon dendrites that support multiplication mediated by conductance-based input in an interception model inspired by the biological dragonfly. We also demonstrate spatiotemporal pattern recognition and direction selectivity using dendrites on the Loihi neuromorphic platform. These dendritic circuits can be assembled hierarchically as building blocks for classifying complex spatiotemporal patterns.

This review investigates the transformative potential of neuromorphic computing in advancing biointegrated electronics, with a particular emphasis on applications in medical sensing, diagnostics, and therapeutic interventions. By examining the convergence of edge computing and neuromorphic principles, we explore how emulating the operational principles of the human brain can enhance the energy efficiency and functionality of biointegrated electronics. The review begins with an introduction to recent breakthroughs in materials and circuit designs that aim to mimic various aspects of the biologi...

This work proves that neuromorphic computing is Turing-complete and therefore capable of general-purpose computing, and establishes the Turing-completeness of neuromorph computing.

Vertical-Cavity Surface-Emitting Lasers for high-speed and energy-efficiency systems for photonic neuromorphic computing yield excellent performance in complex processing tasks whilst benefitting from hardware-friendly implementations and full compatibility with optical communication technologies.

This paper describes a model of neuromorphic computation and state the assumptions that govern the computational complexity of neuromorph algorithms, and presents a theoretical framework to define the computational complexities of a neuromorphic algorithm.

This work presents an overview of early attempts made to study a neuromorphic approach in a space context at the European Space Agency's (ESA) Advanced Concepts Team (ACT).

This survey reviews computing models and hardware platforms of existing neuromorphic computing systems, and introduces neuron and synapse models, followed by the discussion on how they will affect hardware design.

An overview of various neural networks with a focus on building a memristor‐based spike neural network neuromorphic computing system is provided, and an outlook for brain‐like computing is proposed.

If deep learning wants to follow up the victory that deep learning has won and further build a general, efficient and brain-like intelligence, it is suggested to develop a paradigm of neuromorphic computing, which combines architecture, algorithms, circuits and devices tightly.

This work considers a reservoir comprised of a single qudit ($d$-dimensional quantum system), and demonstrates a robust performance advantage compared to an analogous classical system accompanied by a clear improvement with Hilbert space dimension for two benchmark tasks.

This work identifies several science areas where neuromorphic computing can either make an immediate impact or the societal impact would be extremely high if the technological barriers can be addressed.

Halide perovskite is an emerging material with excellent optoelectronic properties, and also widely used in neuromorphic devices. Recently, halide perovskite has been redefined as exhibiting extraordinary multifunction, e.g., photoferroelectricity. Herein, this work employs a composite material consisting of halide perovskite and organic ferroelectric material to develop a new photoferroelectric synapse, and the photoferroelectricity and some synaptic plasticity are investigated. By the corresponding test analysis, it is demonstrated that photoelectricity and ferroelectricity can reinforce eac...

Brain-inspired neuromorphic computing and portable intelligent electronic products have received increasing attention. In the present work, nanocellulose-gated indium tin oxide neuromorphic transistors are fabricated. The device exhibits good electrical performance. Short-term synaptic plasticities were mimicked, including excitatory postsynaptic current, paired-pulse facilitation, and dynamic high-pass synaptic filtering. Interestingly, an effective linear synaptic weight updating strategy was adopted, resulting in an excellent recognition accuracy of ∼92.93% for the Modified National Institu...

A versatile memristor that enables non-volatile memory, selectors, artificial neurons, and artificial synapses, which will provide advantages regarding circuit simplification, fabrication processes, and manufacturing costs is developed.

The progress, challenges, and opportunities for both volatile and nonvolatile memristor in the level of materials, integration technology, algorithm, and system are highlighted in this review.

This perspective article discusses the different implementations of quantum neuromorphic networks with digital and analog circuits, highlight their respective advantages, and review exciting recent experimental results.

A new architectural technique is proposed, by designing an intelligent run-time manager (NCRTM), which dynamically destresses neuron and synapse circuits in response to the short-term aging in their CMOS transistors during the execution of machine learning workloads, with the objective of meeting a reliability target.

The roadmap is organized into several thematic sections, outlining current computing challenges, discussing the neuromorphic computing approach, analyzing mature and currently utilized technologies, providing an overview of emerging technologies, addressing material challenges, exploring novel computing concepts, and finally examining the maturity level of emerging technologies while determining the next essential steps for their advancement.

This roadmap envisages the potential applications of neuromorphic materials in cutting edge technologies and focuses on the design and fabrication of artificial neural systems, which takes inspiration from biology, physics, mathematics, computer science and engineering.

The switching mechanisms that endow memristive devices with rich dynamics and nonlinearity are highlighted, and subsequently various nonlinear spiking neuron behaviors emulated in these memristIVE devices are reviewed.

Many emerging AI applications—especially those that must operate in unpredictable real-world environments with power, latency, and data constraints—require fundamentally new approaches and deep neural networks are needed to address these challenges.

A systematic summary of Boolean logic computing based on emerging neuromorphic transistors is presented and it is believed that comprehensive investigations on neuromorphic Boolean logic operations are crucial to push the development of future neuromorphic computing toward high efficiency and high integration density.

These insights uncover computing principles, primitives, and algorithms on different levels of abstraction and call for more research into the basis of neural computation and neuronally inspired computing hardware.

The dominant mechanisms for reconfigurability are comprehensively concluded, categorized into ion migration, carrier migration, phase transition, spintronics, and photonics, and a perspective on the future challenges for reconfigured neuromorphic computing is discussed, definitely expanding its horizon for scientific communities.

The current progress of NW-based synaptic memristors and synaptic transistors and the challenges faced by NW- based synaptic devices will be proposed and it is hoped this perspective will be beneficial for the application of NW -based synaptic devices in neuromorphic systems.

Integrated multifunctionality in visual information processing is crucial in the artificial intelligence era. Compared to the parallel human vision system, current bionic vision devices exhibit a complex structure with single functionality, challenging intelligent processing and integration. Here, a multisensory artificial synapse with a crossbar structure comprising graphene/α‐In2Se3/graphene layers is demonstrated, merging sensing, memory, and computing while mimicking various synaptic properties. The Schottky barrier height is modulated by the polarization of ferroelectric semiconductor α‐I...

The shift towards a distributed computing paradigm, where multiple systems acquire and elaborate data in real-time, leads to challenges that must be met. In particular, it is becoming increasingly essential to compute on the edge of the network, close to the sensor collecting data. The requirements of a system operating on the edge are very tight: power efficiency, low area occupation, fast response times, and on-line learning. Brain-inspired architectures such as spiking neural networks (SNNs) use artificial neurons and synapses that simultaneously perform low-latency computation and internal...

A reconfigurable hardware platform, which can switch from continuously modulated conductance for emulating synapse to spiking behavior for mimicking neuron, is designed and demonstrated on a spiking neural network with an accuracy of 95.8% and self-adaptive grow-when required network.

This Focus Issue presents advances on various aspects of algorithms for neuromorphic computing, from very fundamental questions about the computational properties of the basic computing elements in neuromorphic systems, algorithms for continual learning, semantic segmentation, and novel efficient learning paradigms, up to algorithms for a specific application domain.

This Perspective discusses select examples of quantum-material-based devices for neuromorphic functionalities into larger emergent complex network systems and provides an outlook on the current opportunities and challenges.

A list of insights distilled from this survey which give general guidelines for the design of future neuromorphic systems are concluded.

The current state-of-the-art for non-cognitive applications on neuromorphic computers, including simple computational kernels for composition, graph algorithms, constrained optimization, and signal processing are reviewed.

Recent advances in WDM-based ONNs are reviewed, focusing on methods that use integrated microcombs to implement ONN's, and results for human image processing using an optical convolution accelerator operating at 11 Tera operations per second are presented.

Brian Gaines observes that computing has been a recursive technology: it supports other technologies that in turn support the progress of computing itself, leading to a positive exponential feedback loop and an exponential growth.

It is shown for a two qubit system that quantum gates can be learned as a change of parameters for neural network dynamics, and the proposal for probabilistic computing goes beyond Markov chains and is not based on transition probabilities.

It is shown that the proposed metric correlates with the per-sample fault coverage and that retaining a set of high-ranked samples in the order of ten achieves near-perfect fault coverage for critical faults that affect the SNN accuracy.

This survey reviews results that are obtained to date with Loihi across the major algorithmic domains under study, including deep learning approaches and novel approaches that aim to more directly harness the key features of spike-based neuromorphic hardware.