Neuromorphic computing has come to refer to a variety of brain-inspired computers, devices, and models that contrast the pervasive von Neumann computer architecture This biologically inspired approach has created highly connected synthetic neurons and synapses that can be used to model neuroscience theories as well as solve challenging machine learning problems The promise of the technology is to create a brain-like ability to learn and adapt, but the technical challenges are significant, starting with an accurate neuroscience model of how the brain works, to finding materials and engineering breakthroughs to build devices to support these models, to creating a programming framework so the systems can learn, to creating applications with brain-like capabilities In this work, we provide a comprehensive survey of the research and motivations for neuromorphic computing over its history We begin with a 35-year review of the motivations and drivers of neuromorphic computing, then look at the major research areas of the field, which we define as neuro-inspired models, algorithms and learning approaches, hardware and devices, supporting systems, and finally applications We conclude with a broad discussion on the major research topics that need to be addressed in the coming years to see the promise of neuromorphic computing fulfilled The goals of this work are to provide an exhaustive review of the research conducted in neuromorphic computing since the inception of the term, and to motivate further work by illuminating gaps in the field where new research is needed

A Survey of Neuromorphic Computing and Neural Networks in Hardware.

The rapid development of information technology has led to urgent requirements for high efficiency and ultralow power consumption. In the past few decades, neuromorphic computing has drawn extensive attention due to its promising capability in processing massive data with extremely low power consumption. Here, we offer a comprehensive review on emerging artificial neuromorphic devices and their applications. In light of the inner physical processes, we classify the devices into nine major categories and discuss their respective strengths and weaknesses. We will show that anion/cation migration-based memristive devices, phase change, and spintronic synapses have been quite mature and possess excellent stability as a memory device, yet they still suffer from challenges in weight updating linearity and symmetry. Meanwhile, the recently developed electrolyte-gated synaptic transistors have demonstrated outstanding energy efficiency, linearity, and symmetry, but their stability and scalability still need to be optimized. Other emerging synaptic structures, such as ferroelectric, metal–insulator transition based, photonic, and purely electronic devices also have limitations in some aspects, therefore leading to the need for further developing high-performance synaptic devices. Additional efforts are also demanded to enhance the functionality of artificial neurons while maintaining a relatively low cost in area and power, and it will be of significance to explore the intrinsic neuronal stochasticity in computing and optimize their driving capability, etc. Finally, by looking into the correlations between the operation mechanisms, material systems, device structures, and performance, we provide clues to future material selections, device designs, and integrations for artificial synapses and neurons.

A comprehensive review on emerging artificial neuromorphic devices

As the research on artificial intelligence booms, there is broad interest in brain-inspired computing using novel neuromorphic devices. The potential of various emerging materials and devices for neuromorphic computing has attracted extensive research efforts, leading to a large number of publications. Going forward, in order to better emulate the brain's functions, its relevant fundamentals, working mechanisms, and resultant behaviors need to be re-visited, better understood, and connected to electronics. A systematic overview of biological and artificial neural systems is given, along with their related critical mechanisms. Recent progress in neuromorphic devices is reviewed and, more importantly, the existing challenges are highlighted to hopefully shed light on future research directions.

Bridging Biological and Artificial Neural Networks with Emerging Neuromorphic Devices: Fundamentals, Progress, and Challenges.

The existence of reversible resistance switching (RS) behaviors induced by electric stimulus has been known for some time, and these intriguing physical phenomena have been observed in numerous materials, including oxides. As conventional charge-based random access memory is expected to face a size limit in the near future, a surge of renewed interest has been developed in RS phenomena for possible applications in small nonvolatile memory devices called resistance random access memory (RRAM). Of particular interest is unipolar RS, which shows the RS at two values of applied voltage of the same polarity. The unipolar RS exhibits a much larger resistance change than other RS phenomena, and this greatly simplifies the process of reading the memory state. When fabricated with oxide p-n diodes, memory cells using unipolar RS can be stacked vertically, which has the potential for dramatically increasing memory density. Therefore, unipolar RRAM may be a good candidate for multi-stacked, high density, nonvolatile memory. The most important scientific and technical issues concerning unipolar RS are how it works and the identification of its controlling parameters. Some studies have reported that unipolar RS comes from a homogeneous/inhomogeneous transition of current distribution, while others maintain that it comes from the formation and rupture of conducting filaments. Even with recent extensive studies on unipolar RS, its basic origin is still far from being understood. In addition, no model exists that actually explains how the reversible switching can occur at two values of applied voltage. This lack of a quantitative model poses a major barrier for unipolar RRAM applications. In this study, we describe RS behavior in polycrystalline TiO2 film. To explain the basic mechanism of unipolar RS behavior, we propose a new percolation model based on a network of ‘‘circuit breakers’’ with two switchable metastable states. The random circuit breaker (RCB) network model can explain the long-standing material issue of how unipolar RS occurs. This simple percolation model is different from the conventional percolation models, which have dealt only with static or irreversible dynamic processes. In addition, the RCB network model provides an indication of how to overcome the substantial distribution of switching voltages, which is currently considered the most serious obstacle to practical unipolar RRAM applications. The unipolar RS phenomenon can be explained by the current (I)-voltage (V) curves in Figure 1a, which are derived from measurements of our polycrystalline TiO2 thin capacitors. At the pristine state (green dot), they are in an insulating state. As the external voltage Vext increases from zero and reaches a threshold voltage Vforming, a sudden increase occurs in the current. If the current is not limited to a certain value, here called the compliance current Icomp, the TiO2 capacitor would experience a dielectric breakdown and be destroyed. However, [*] Prof. T. W. Noh, S. C. Chae, S. B. Lee, S. H Chang, Dr. C. Liu ReCOE & FPRD, Department of Physics and Astronomy Seoul National University Seoul 151-747 (Korea) E-mail: twnoh@snu.ac.kr

Random circuit breaker network model for unipolar resistance switching (Advanced Materials (2008) 20 (1154))

Electronic synaptic devices are important building blocks for neuromorphic computational systems that can go beyond the constraints of von Neumann architecture. Although two-terminal memristive devices are demonstrated to be possible candidates, they suffer from several shortcomings related to the filament formation mechanism including nonlinear switching, write noise, and high device conductance, all of which limit the accuracy and energy efficiency. Electrochemical three-terminal transistors, in which the channel conductance can be tuned without filament formation provide an alternative platform for synaptic electronics. Here, an all-solid-state electrochemical transistor made with Li ion–based solid dielectric and 2D α-phase molybdenum oxide (α-MoO3) nanosheets as the channel is demonstrated. These devices achieve nonvolatile conductance modulation in an ultralow conductance regime (<75 nS) by reversible intercalation of Li ions into the α-MoO3 lattice. Based on this operating mechanism, the essential functionalities of synapses, such as shortand long-term synaptic plasticity and bidirectional near-linear analog weight update are demonstrated. Simulations using the handwritten digit data sets demonstrate high recognition accuracy (94.1%) of the synaptic transistor arrays. These results provide an insight into the application of 2D oxides for large-scale, energy-efficient neuromorphic computing networks.

All‐Solid‐State Synaptic Transistor with Ultralow Conductance for Neuromorphic Computing

Synaptic Suppression Triplet-STDP Learning Rule Realized in Second-Order Memristors

Memory materials and devices: From concept to application

The flexible construction of electrode materials performs important functions in significantly enhancing the electrochemical properties of both energy storage and conversion devices. There is an increasing demand, not only for the current, but also the near-future energy field, where both high energy and power densities are necessary for the fabricated electrode materials. Carbon based nanoporous electrode materials demonstrate diverse superiority with respect to their application in the electrochemical capacitors (i.e., supercapacitors) industry. The striking characteristics of mesoporous carbons include abundance in raw materials, sustainability, good conductivity, excellent heat and chemical resistance, good workability, and lightweight nanostructure. Recently, various mesoporous carbons have increasingly been applied in the energy storage field because of their intrinsic open pore structure as well as high surface areas. However, the relationships between the synthesis strategies and the product struc...

Nanocasting and Direct Synthesis Strategies for Mesoporous Carbons as Supercapacitor Electrodes

A Review of Resistive Switching Devices: Performance Improvement, Characterization, and Applications

The sensory nervous system (SNS) builds up the association between external stimuli and the response of organisms. In this system, habituation is a fundamental characteristic that filters out irrelevantly repetitive information and makes the SNS adapt to the external environment. To emulate this critical process in electronic devices, a Lix SiOy -based memristor (TiN/Lix SiOy /Pt) is developed where the temporal response under repetitive stimulation is similar to that of habituation. By connecting this synaptic device to a leaky integrate-and-fire neuron based on a Ag/SiO2 :Ag/Au memristor, a fully memristive SNS with habituation is experimentally demonstrated. Finally, a habituation spiking neural network based on the SNS is built and its application in obstacle avoidance for robot navigation is successfully presented. The results provide that a direct emulation of the biologically inspired learning process by memristors could be a sound choice for neuromorphic hardware implementation.

Tuo Shi

Papers

Synaptic Suppression Triplet-STDP Learning Rule Realized in Second-Order Memristors

Memory materials and devices: From concept to application

Nanocasting and Direct Synthesis Strategies for Mesoporous Carbons as Supercapacitor Electrodes

A Review of Resistive Switching Devices: Performance Improvement, Characterization, and Applications

A Habituation Sensory Nervous System with Memristors