Joseph S. Friedman

Gleb Krylov, Alexander J. Edwards, Joseph S. Friedman et al.

Conventional semiconductor-based integrated circuits are gradually approaching fundamental scaling limits. Many prospective solutions have recently emerged to supplement or replace both the technology on which basic devices are built and the architecture of data processing. Neuromorphic circuits are a promising approach to computing where techniques used by the brain to achieve high efficiency are exploited. Many existing neuromorphic circuits rely on unconventional and useful properties of novel technologies to better mimic the operation of the brain. One such technology is single flux quantum (SFQ) logic -- a cryogenic superconductive technology in which the data are represented by quanta of magnetic flux (fluxons) produced and processed by Josephson junctions embedded within inductive loops. The movement of a fluxon within a circuit produces a quantized voltage pulse (SFQ pulse), resembling a neuronal spiking event. These circuits routinely operate at clock frequencies of tens to hundreds of gigahertz, making SFQ a natural technology for processing high frequency pulse trains. Prior proposals for SFQ neural networks often require energy-expensive fluxon conversions, involve heterogeneous technologies, or exclusively focus on device level behavior. In this paper, a design methodology for deep single flux quantum neuromorphic networks is presented. Synaptic and neuronal circuits based on SFQ technology are presented and characterized. Based on these primitives, a deep neuromorphic XOR network is evaluated as a case study, both at the architectural and circuit levels, achieving wide classification margins. The proposed methodology does not employ unconventional superconductive devices or semiconductor transistors. The resulting networks are tunable by an external current, making this proposed system an effective approach for scalable cryogenic neuromorphic computing.

Xi Li, Disha Biswas, Peng Zhou et al.

While software implementations of neural networks have driven significant advances in computation, the von Neumann architecture imposes fundamental limitations on speed and energy efficiency. Neuromorphic networks, with structures inspired by the brain's architecture, offer a compelling solution with the potential to approach the extreme energy efficiency of neurobiological systems. Photonic neuromorphic networks (PNNs) are particularly attractive because they leverage the inherent advantages of light, namely high parallelism, low latency, and exceptional energy efficiency. Previous PNN demonstrations have largely focused on device-level functionalities or system-level implementations reliant on supervised learning and inefficient optical-electrical-optical (OEO) conversions. Here, we introduce a purely photonic deep PNN architecture that enables online, unsupervised learning. We propose a local feedback mechanism operating entirely in the optical domain that implements a Hebbian learning rule using non-volatile phase-change material synapses. We experimentally demonstrate this approach on a non-trivial letter recognition task using a commercially available fiber-optic platform and achieve a 100 percent recognition rate, showcasing an all-optical solution for efficient, real-time information processing. This work unlocks the potential of photonic computing for complex artificial intelligence applications by enabling direct, high-throughput processing of optical information without intermediate OEO signal conversions.

Peng Zhou, Julie A. Smith, Laura Deremo et al.

The use of analog resistance states for storing weights in neuromorphic systems is impeded by fabrication imprecision and device stochasticity that limit the precision of synapse weights. This challenge can be resolved by emulating analog behavior with the stochastic switching of the binary states of spin-transfer torque magnetoresistive random-access memory (STT-MRAM). However, previous approaches based on STT-MRAM operate in an asynchronous manner that is difficult to implement experimentally. This paper proposes a synchronous spiking neural network system with clocked circuits that perform unsupervised learning leveraging the stochastic switching of STT-MRAM. The proposed system enables a single-layer network to achieve 90% inference accuracy on the MNIST dataset.

Peng Zhou, Alexander J. Edwards, Fred B. Mancoff et al.

We present the first experimental demonstration of a neuromorphic network with magnetic tunnel junction (MTJ) synapses, which performs image recognition via vector-matrix multiplication. We also simulate a large MTJ network performing MNIST handwritten digit recognition, demonstrating that MTJ crossbars can match memristor accuracy while providing increased precision, stability, and endurance.

Thomas Leonard, Samuel Liu, Mahshid Alamdar et al.

In neuromorphic computing, artificial synapses provide a multi-weight conductance state that is set based on inputs from neurons, analogous to the brain. Additional properties of the synapse beyond multiple weights can be needed, and can depend on the application, requiring the need for generating different synapse behaviors from the same materials. Here, we measure artificial synapses based on magnetic materials that use a magnetic tunnel junction and a magnetic domain wall. By fabricating lithographic notches in a domain wall track underneath a single magnetic tunnel junction, we achieve 4-5 stable resistance states that can be repeatably controlled electrically using spin orbit torque. We analyze the effect of geometry on the synapse behavior, showing that a trapezoidal device has asymmetric weight updates with high controllability, while a straight device has higher stochasticity, but with stable resistance levels. The device data is input into neuromorphic computing simulators to show the usefulness of application-specific synaptic functions. Implementing an artificial neural network applied on streamed Fashion-MNIST data, we show that the trapezoidal magnetic synapse can be used as a metaplastic function for efficient online learning. Implementing a convolutional neural network for CIFAR-100 image recognition, we show that the straight magnetic synapse achieves near-ideal inference accuracy, due to the stability of its resistance levels. This work shows multi-weight magnetic synapses are a feasible technology for neuromorphic computing and provides design guidelines for emerging artificial synapse technologies.

Pranav O. Mathews, Christian B. Duffee, Abel Thayil et al.

Neuromorphic computing systems overcome the limitations of traditional von Neumann computing architectures. These computing systems can be further improved upon by using emerging technologies that are more efficient than CMOS for neural computation. Recent research has demonstrated memristors and spintronic devices in various neural network designs boost efficiency and speed. This paper presents a biologically inspired fully spintronic neuron used in a fully spintronic Hopfield RNN. The network is used to solve tasks, and the results are compared against those of current Hopfield neuromorphic architectures which use emerging technologies.

Alexander J. Edwards, Dhritiman Bhattacharya, Peng Zhou et al.

Reservoir computing (RC) has received recent interest because reservoir weights do not need to be trained, enabling extremely low-resource consumption implementations, which could have a transformative impact on edge computing and in-situ learning where resources are severely constrained. Ideally, a natural hardware reservoir should be passive, minimal, expressive, and feasible; to date, proposed hardware reservoirs have had difficulty meeting all of these criteria. We therefore propose a reservoir that meets all of these criteria by leveraging the passive interactions of dipole-coupled, frustrated nanomagnets. The frustration significantly increases the number of stable reservoir states, enriching reservoir dynamics, and as such these frustrated nanomagnets fulfill all of the criteria for a natural hardware reservoir. We likewise propose a complete frustrated nanomagnet reservoir computing (NMRC) system with low-power complementary metal-oxide semiconductor (CMOS) circuitry to interface with the reservoir, and initial experimental results demonstrate the reservoir's feasibility. The reservoir is verified with micromagnetic simulations on three separate tasks demonstrating expressivity. The proposed system is compared with a CMOS echo-state-network (ESN), demonstrating an overall resource decrease by a factor of over 10,000,000, demonstrating that because NMRC is naturally passive and minimal it has the potential to be extremely resource efficient.

Samuel Liu, Christopher H. Bennett, Joseph S. Friedman et al.

Neuromorphic computing with spintronic devices has been of interest due to the limitations of CMOS-driven von Neumann computing. Domain wall-magnetic tunnel junction (DW-MTJ) devices have been shown to be able to intrinsically capture biological neuron behavior. Edgy-relaxed behavior, where a frequently firing neuron experiences a lower action potential threshold, may provide additional artificial neuronal functionality when executing repeated tasks. In this study, we demonstrate that this behavior can be implemented in DW-MTJ artificial neurons via three alternative mechanisms: shape anisotropy, magnetic field, and current-driven soft reset. Using micromagnetics and analytical device modeling to classify the Optdigits handwritten digit dataset, we show that edgy-relaxed behavior improves both classification accuracy and classification rate for ordered datasets while sacrificing little to no accuracy for a randomized dataset. This work establishes methods by which artificial spintronic neurons can be flexibly adapted to datasets.

Wesley H. Brigner, Naimul Hassan, Xuan Hu et al.

Complementary metal oxide semiconductor (CMOS) devices display volatile characteristics, and are not well suited for analog applications such as neuromorphic computing. Spintronic devices, on the other hand, exhibit both non-volatile and analog features, which are well-suited to neuromorphic computing. Consequently, these novel devices are at the forefront of beyond-CMOS artificial intelligence applications. However, a large quantity of these artificial neuromorphic devices still require the use of CMOS, which decreases the efficiency of the system. To resolve this, we have previously proposed a number of artificial neurons and synapses that do not require CMOS for operation. Although these devices are a significant improvement over previous renditions, their ability to enable neural network learning and recognition is limited by their intrinsic activation functions. This work proposes modifications to these spintronic neurons that enable configuration of the activation functions through control of the shape of a magnetic domain wall track. Linear and sigmoidal activation functions are demonstrated in this work, which can be extended through a similar approach to enable a wide variety of activation functions.

Alvaro Velasquez, Christopher H. Bennett, Naimul Hassan et al.

We propose a hardware learning rule for unsupervised clustering within a novel spintronic computing architecture. The proposed approach leverages the three-terminal structure of domain-wall magnetic tunnel junction devices to establish a feedback loop that serves to train such devices when they are used as synapses in a neuromorphic computing architecture.

Peng Zhou, Nathan R. McDonald, Alexander J. Edwards et al.

Reservoir computing is an emerging methodology for neuromorphic computing that is especially well-suited for hardware implementations in size, weight, and power (SWaP) constrained environments. This work proposes a novel hardware implementation of a reservoir computer using a planar nanomagnet array. A small nanomagnet reservoir is demonstrated via micromagnetic simulations to be able to identify simple waveforms with 100% accuracy. Planar nanomagnet reservoirs are a promising new solution to the growing need for dedicated neuromorphic hardware.

Christopher H. Bennett, T. Patrick Xiao, Can Cui et al.

Machine learning implements backpropagation via abundant training samples. We demonstrate a multi-stage learning system realized by a promising non-volatile memory device, the domain-wall magnetic tunnel junction (DW-MTJ). The system consists of unsupervised (clustering) as well as supervised sub-systems, and generalizes quickly (with few samples). We demonstrate interactions between physical properties of this device and optimal implementation of neuroscience-inspired plasticity learning rules, and highlight performance on a suite of tasks. Our energy analysis confirms the value of the approach, as the learning budget stays below 20 $μJ$ even for large tasks used typically in machine learning.

Wesley H. Brigner, Naimul Hassan, Xuan Hu et al.

Neuromorphic computing promises revolutionary improvements over conventional systems for applications that process unstructured information. To fully realize this potential, neuromorphic systems should exploit the biomimetic behavior of emerging nanodevices. In particular, exceptional opportunities are provided by the non-volatility and analog capabilities of spintronic devices. While spintronic devices have previously been proposed that emulate neurons and synapses, complementary metal-oxide-semiconductor (CMOS) devices are required to implement multilayer spintronic perceptron crossbars. This work therefore proposes a new spintronic neuron that enables purely spintronic multilayer perceptrons, eliminating the need for CMOS circuitry and simplifying fabrication.

Farid Kenarangi, Xuan Hu, Yihan Liu et al.

Ambipolar carbon nanotube based field-effect transistors (AP-CNFETs) exhibit unique electrical characteristics, such as tri-state operation and bi-directionality, enabling systems with complex and reconfigurable computing. In this paper, AP-CNFETs are used to design a mixed-signal machine learning (ML) classifier. The classifier is designed in SPICE with feature size of 15 nm and operates at 250 MHz. The system is demonstrated based on MNIST digit dataset, yielding 90% accuracy and no accuracy degradation as compared with the classification of this dataset in Python. The system also exhibits lower power consumption and smaller physical size as compared with the state-of-the-art CMOS and memristor based mixed-signal classifiers.