Tau-Cell-Based Analog Silicon Retina With Spatio- Temporal Filtering and Contrast Gain Control.

Developing precise artificial retinas is crucial because they hold the potential to restore vision, improve visual prosthetics, and enhance computer vision systems. Emulating the luminance and contrast adaption features of the retina is essential to improve visual perception and efficiency to provide an environment realistic representation to the user. In this article, we introduce an artificial retina model that leverages its potent adaptation to luminance and contrast to enhance vision sensing and information processing. The model has the ability to achieve the realization of both tonic and phasic cells in the simplest manner. We have implemented the retina model using 0.18 µm process technology and validated the accuracy of the hardware implementation through circuit simulation that closely matches the software retina model. Additionally, we have characterized a single pixel fabricated using the same 0.18 µm process. This pixel demonstrates an 87.7-% ratio of variance with the temporal software model and operates with a power consumption of 369 nW.

Event-driven spectrotemporal feature extraction and classification using a silicon cochlea model.

This paper presents a reconfigurable digital implementation of an event-based binaural cochlear system on a Field Programmable Gate Array (FPGA). It consists of a pair of the Cascade of Asymmetric Resonators with Fast Acting Compression (CAR-FAC) cochlea models and leaky integrate-and-fire (LIF) neurons. Additionally, we propose an event-driven SpectroTemporal Receptive Field (STRF) Feature Extraction using Adaptive Selection Thresholds (FEAST). It is tested on the TIDIGTIS benchmark and compared with current event-based auditory signal processing approaches and neural networks.

Real-Time Event-Based Unsupervised Feature Consolidation and Tracking for Space Situational Awareness.

Earth orbit is a limited natural resource that hosts a vast range of vital space-based systems that support the international community's national, commercial and defence interests. This resource is rapidly becoming depleted with over-crowding in high demand orbital slots and a growing presence of space debris. We propose the Fast Iterative Extraction of Salient targets for Tracking Asynchronously (FIESTA) algorithm as a robust, real-time and reactive approach to optical Space Situational Awareness (SSA) using Event-Based Cameras (EBCs) to detect, localize, and track Resident Space Objects (RSOs) accurately and timely. We address the challenges of the asynchronous nature and high temporal resolution output of the EBC accurately, unsupervised and with few tune-able parameters using concepts established in the neuromorphic and conventional tracking literature. We show this algorithm is capable of highly accurate in-frame RSO velocity estimation and average sub-pixel localization in a simulated test environment to distinguish the capabilities of the EBC and optical setup from the proposed tracking system. This work is a fundamental step toward accurate end-to-end real-time optical event-based SSA, and developing the foundation for robust closed-form tracking evaluated using standardized tracking metrics.

Martingales and the fixation time of evolutionary graphs with arbitrary dimensionality.

Evolutionary graph theory (EGT) investigates the Moran birth-death process constrained by graphs. Its two principal goals are to find the fixation probability and time for some initial population of mutants on the graph. The fixation probability of graphs has received considerable attention. Less is known about the distribution of fixation time. We derive clean, exact expressions for the full conditional characteristic functions (CCFs) of a close proxy to fixation and extinction times. That proxy is the number of times that the mutant population size changes before fixation or extinction. We derive these CCFs from a product martingale that we identify for an evolutionary graph with any number of partitions. The existence of that martingale only requires that the connections between those partitions are of a certain type. Our results are the first expressions for the CCFs of any proxy to fixation time on a graph with any number of partitions. The parameter dependence of our CCFs is explicit, so we can explore how they depend on graph structure. Martingales are a powerful approach to study principal problems of EGT. Their applicability is invariant to the number of partitions in a graph, so we can study entire families of graphs simultaneously.

Neuromorphic Engineering Needs Closed-Loop Benchmarks.

Neuromorphic engineering aims to build (autonomous) systems by mimicking biological systems. It is motivated by the observation that biological organisms-from algae to primates-excel in sensing their environment, reacting promptly to their perils and opportunities. Furthermore, they do so more resiliently than our most advanced machines, at a fraction of the power consumption. It follows that the performance of neuromorphic systems should be evaluated in terms of real-time operation, power consumption, and resiliency to real-world perturbations and noise using task-relevant evaluation metrics. Yet, following in the footsteps of conventional machine learning, most neuromorphic benchmarks rely on recorded datasets that foster sensing accuracy as the primary measure for performance. Sensing accuracy is but an arbitrary proxy for the actual system's goal-taking a good decision in a timely manner. Moreover, static datasets hinder our ability to study and compare closed-loop sensing and control strategies that are central to survival for biological organisms. This article makes the case for a renewed focus on closed-loop benchmarks involving real-world tasks. Such benchmarks will be crucial in developing and progressing neuromorphic Intelligence. The shift towards dynamic real-world benchmarking tasks should usher in richer, more resilient, and robust artificially intelligent systems in the future.

Noise-robust text-dependent speaker identification using cochlear models.

One challenging issue in speaker identification (SID) is to achieve noise-robust performance. Humans can accurately identify speakers, even in noisy environments. We can leverage our knowledge of the function and anatomy of the human auditory pathway to design SID systems that achieve better noise-robust performance than conventional approaches. We propose a text-dependent SID system based on a real-time cochlear model called cascade of asymmetric resonators with fast-acting compression (CARFAC). We investigate the SID performance of CARFAC on signals corrupted by noise of various types and levels. We compare its performance with conventional auditory feature generators including mel-frequency cepstrum coefficients, frequency domain linear predictions, as well as another biologically inspired model called the auditory nerve model. We show that CARFAC outperforms other approaches when signals are corrupted by noise. Our results are consistent across datasets, types and levels of noise, different speaking speeds, and back-end classifiers. We show that the noise-robust SID performance of CARFAC is largely due to its nonlinear processing of auditory input signals. Presumably, the human auditory system achieves noise-robust performance via inherent nonlinearities as well.

Martingales and the characteristic functions of absorption time on bipartite graphs.

Evolutionary graph theory investigates how spatial constraints affect processes that model evolutionary selection, e.g. the Moran process. Its principal goals are to find the fixation probability and the conditional distributions of fixation time, and show how they are affected by different graphs that impose spatial constraints. Fixation probabilities have generated significant attention, but much less is known about the conditional time distributions, even for simple graphs. Those conditional time distributions are difficult to calculate, so we consider a close proxy to it: the number of times the mutant population size changes before absorption. We employ martingales to obtain the conditional characteristic functions (CCFs) of that proxy for the Moran process on the complete bipartite graph. We consider the Moran process on the complete bipartite graph as an absorbing random walk in two dimensions. We then extend Wald's martingale approach to sequential analysis from one dimension to two. Our expressions for the CCFs are novel, compact, exact, and their parameter dependence is explicit. We show that our CCFs closely approximate those of absorption time. Martingales provide an elegant framework to solve principal problems of evolutionary graph theory. It should be possible to extend our analysis to more complex graphs than we show here.

Event Camera Simulator Improvements via Characterized Parameters.

It has been more than two decades since the first neuromorphic Dynamic Vision Sensor (DVS) sensor was invented, and many subsequent prototypes have been built with a wide spectrum of applications in mind. Competing against state-of-the-art neural networks in terms of accuracy is difficult, although there are clear opportunities to outperform conventional approaches in terms of power consumption and processing speed. As neuromorphic sensors generate sparse data at the focal plane itself, they are inherently energy-efficient, data-driven, and fast. In this work, we present an extended DVS pixel simulator for neuromorphic benchmarks which simplifies the latency and the noise models. In addition, to more closely model the behaviour of a real pixel, the readout circuitry is modelled, as this can strongly affect the time precision of events in complex scenes. Using a dynamic variant of the MNIST dataset as a benchmarking task, we use this simulator to explore how the latency of the sensor allows it to outperform conventional sensors in terms of sensing speed.

Correction to 'Wald's martingale and the conditional distributions of absorption time in the Moran process'.

[This corrects the article DOI: 10.1098/rspa.2020.0135.].

Event-Based Computation for Touch Localization Based on Precise Spike Timing.

Precise spike timing and temporal coding are used extensively within the nervous system of insects and in the sensory periphery of higher order animals. However, conventional Artificial Neural Networks (ANNs) and machine learning algorithms cannot take advantage of this coding strategy, due to their rate-based representation of signals. Even in the case of artificial Spiking Neural Networks (SNNs), identifying applications where temporal coding outperforms the rate coding strategies of ANNs is still an open challenge. Neuromorphic sensory-processing systems provide an ideal context for exploring the potential advantages of temporal coding, as they are able to efficiently extract the information required to cluster or classify spatio-temporal activity patterns from relative spike timing. Here we propose a neuromorphic model inspired by the sand scorpion to explore the benefits of temporal coding, and validate it in an event-based sensory-processing task. The task consists in localizing a target using only the relative spike timing of eight spatially-separated vibration sensors. We propose two different approaches in which the SNNs learns to cluster spatio-temporal patterns in an unsupervised manner and we demonstrate how the task can be solved both analytically and through numerical simulation of multiple SNN models. We argue that the models presented are optimal for spatio-temporal pattern classification using precise spike timing in a task that could be used as a standard benchmark for evaluating event-based sensory processing models based on temporal coding.

Low Power, CMOS-MoS2 Memtransistor based Neuromorphic Hybrid Architecture for Wake-Up Systems.

Neuromorphic architectures have become essential building blocks for next-generation computational systems, where intelligence is embedded directly onto low power, small area, and computationally efficient hardware devices. In such devices, realization of neural algorithms requires storage of weights in digital memories, which is a bottleneck in terms of power and area. We hereby propose a biologically inspired low power, hybrid architectural framework for wake-up systems. This architecture utilizes our novel high-performance, ultra-low power molybdenum disulphide (MoS2) based two-dimensional synaptic memtransistor as an analogue memory. Furthermore, it exploits random device mismatches to implement the population coding scheme. Power consumption per CMOS neuron block was found to be 3 nw in the 65 nm process technology, while the energy consumption per cycle was 0.3 pJ for potentiation and 20 pJ for depression cycles of the synaptic device. The proposed framework was demonstrated for classification and regression tasks, using both off-chip and simplified on-chip sign-based learning techniques.

Vibrotactile sensitivity of patients with HIV-related sensory neuropathy: An exploratory study.

BACKGROUND: HIV-associated distal polyneuropathy (HIV-PN) affects large and small sensory nerve fibers and can cause tactile insensitivity. This exploratory study forms part of an effort to apply subsensory electrical nerve stimulation (SENS) to improve tactile sensitivity of patients with HIV-PN. This work presented an opportunity to use a robust protocol to quantitatively describe the vibrotactile sensitivity of individuals with HIV-PN on effective antiretroviral therapy (ART) and correlate these findings with commonly used clinical vibration testing and scoring grades. METHODS: The vibration perception thresholds (VPTs) of 20 patients with HIV-PN at three vibration frequencies (25, 50, and 128 Hz) were measured. We compare the vibration perception threshold (VPT) outcomes to an age- and gender-matched control cohort. We further correlated VPT findings with 128 Hz tuning fork (TF) assessments performed on the HIV-PN participants, accrued as part of a larger study. HIV-PN was defined as having at least one distal symmetrical neuropathic sign, although 18 of 20 had at least two neuropathic signs. CONCLUSIONS: HIV-PN participants were found to have lower VPT sensitivity than controls for all three vibration frequencies, and VPT was more sensitive at higher vibration frequencies for both HIV-PN and controls. VPT sensitivity was reduced with older age. Years on ART was correlated with VPT-25 Hz but not with VPT in general. Notably, VPT sensitivity did not correlate with the clinically used 128 Hz TF severity grades. Outcomes of tests for interaction with vibration frequency suggest that HIV-PN pathology does not affect all mechanoreceptors similarly.

Large-Scale Neuromorphic Spiking Array Processors: A Quest to Mimic the Brain.

Neuromorphic engineering (NE) encompasses a diverse range of approaches to information processing that are inspired by neurobiological systems, and this feature distinguishes neuromorphic systems from conventional computing systems. The brain has evolved over billions of years to solve difficult engineering problems by using efficient, parallel, low-power computation. The goal of NE is to design systems capable of brain-like computation. Numerous large-scale neuromorphic projects have emerged recently. This interdisciplinary field was listed among the top 10 technology breakthroughs of 2014 by the MIT Technology Review and among the top 10 emerging technologies of 2015 by the World Economic Forum. NE has two-way goals: one, a scientific goal to understand the computational properties of biological neural systems by using models implemented in integrated circuits (ICs); second, an engineering goal to exploit the known properties of biological systems to design and implement efficient devices for engineering applications. Building hardware neural emulators can be extremely useful for simulating large-scale neural models to explain how intelligent behavior arises in the brain. The principal advantages of neuromorphic emulators are that they are highly energy efficient, parallel and distributed, and require a small silicon area. Thus, compared to conventional CPUs, these neuromorphic emulators are beneficial in many engineering applications such as for the porting of deep learning algorithms for various recognitions tasks. In this review article, we describe some of the most significant neuromorphic spiking emulators, compare the different architectures and approaches used by them, illustrate their advantages and drawbacks, and highlight the capabilities that each can deliver to neural modelers. This article focuses on the discussion of large-scale emulators and is a continuation of a previous review of various neural and synapse circuits (Indiveri et al., 2011). We also explore applications where these emulators have been used and discuss some of their promising future applications.

Breaking Liebig's Law: An Advanced Multipurpose Neuromorphic Engine.

We present a massively-parallel scalable multi-purpose neuromorphic engine. All existing neuromorphic hardware systems suffer from Liebig's law (that the performance of the system is limited by the component in shortest supply) as they have fixed numbers of dedicated neurons and synapses for specific types of plasticity. For any application, it is always the availability of one of these components that limits the size of the model, leaving the others unused. To overcome this problem, our engine adopts a unique novel architecture: an array of identical components, each of which can be configured as a leaky-integrate-and-fire (LIF) neuron, a learning-synapse, or an axon with trainable delay. Spike timing dependent plasticity (STDP) and spike timing dependent delay plasticity (STDDP) are the two supported learning rules. All the parameters are stored in the SRAMs such that runtime reconfiguration is supported. As a proof of concept, we have implemented a prototype system with 16 neural engines, each of which consists of 32768 (32k) components, yielding half a million components, on an entry level FPGA (Altera Cyclone V). We verified the prototype system with measurement results. To demonstrate that our neuromorphic engine is a high performance and scalable digital design, we implemented it using TSMC 28nm HPC technology. Place and route results using Cadence Innovus with a clock frequency of 2.5 GHz show that this engine achieves an excellent area efficiency of 1.68 µm2 per component: 256k (218) components in a silicon area of 650 µm × 680 µm (∼0.44 mm2, the utilization of the silicon area is 98.7%). The power consumption of this engine is 37 mW, yielding a power efficiency of 0.92 pJ per synaptic operation (SOP).

Astrocytic modulation of cortical oscillations.

Brain waves are rhythmic voltage oscillations emerging from the synchronization of individual neurons into a neuronal network. These oscillations range from slow to fast fluctuations, and are classified by power and frequency band, with different frequency bands being associated with specific behaviours. It has been postulated that at least ten distinct mechanisms are required to cover the frequency range of neural oscillations, however the mechanisms that gear the transition between distinct oscillatory frequencies are unknown. In this study, we have used electrophysiological recordings to explore the involvement of astrocytic K+ clearance processes in modulating neural oscillations at both network and cellular levels. Our results indicate that impairment of astrocytic K+ clearance capabilities, either through blockade of K+ uptake or astrocytic connectivity, enhance network excitability and form high power network oscillations over a wide range of frequencies. At the cellular level, local increases in extracellular K+ results in modulation of the oscillatory behaviour of individual neurons, which underlies the network behaviour. Since astrocytes are central for maintaining K+ homeostasis, our study suggests that modulation of their inherent capabilities to clear K+ from the extracellular milieu is a potential mechanism to optimise neural resonance behaviour and thus tune neural oscillations.

Spatial and Temporal Downsampling in Event-Based Visual Classification.

As the interest in event-based vision sensors for mobile and aerial applications grows, there is an increasing need for high-speed and highly robust algorithms for performing visual tasks using event-based data. As event rate and network structure have a direct impact on the power consumed by such systems, it is important to explore the efficiency of the event-based encoding used by these sensors. The work presented in this paper represents the first study solely focused on the effects of both spatial and temporal downsampling on event-based vision data and makes use of a variety of data sets chosen to fully explore and characterize the nature of downsampling operations. The results show that both spatial downsampling and temporal downsampling produce improved classification accuracy and, additionally, a lower overall data rate. A finding is particularly relevant for bandwidth and power constrained systems. For a given network containing 1000 hidden layer neurons, the spatially downsampled systems achieved a best case accuracy of 89.38% on N-MNIST as opposed to 81.03% with no downsampling at the same hidden layer size. On the N-Caltech101 data set, the downsampled system achieved a best case accuracy of 18.25%, compared with 7.43% achieved with no downsampling. The results show that downsampling is an important preprocessing technique in event-based visual processing, especially for applications sensitive to power consumption and transmission bandwidth.

A FPGA Implementation of the CAR-FAC Cochlear Model.

This paper presents a digital implementation of the Cascade of Asymmetric Resonators with Fast-Acting Compression (CAR-FAC) cochlear model. The CAR part simulates the basilar membrane's (BM) response to sound. The FAC part models the outer hair cell (OHC), the inner hair cell (IHC), and the medial olivocochlear efferent system functions. The FAC feeds back to the CAR by moving the poles and zeros of the CAR resonators automatically. We have implemented a 70-section, 44.1 kHz sampling rate CAR-FAC system on an Altera Cyclone V Field Programmable Gate Array (FPGA) with 18% ALM utilization by using time-multiplexing and pipeline parallelizing techniques and present measurement results here. The fully digital reconfigurable CAR-FAC system is stable, scalable, easy to use, and provides an excellent input stage to more complex machine hearing tasks such as sound localization, sound segregation, speech recognition, and so on.

An FPGA-Based Massively Parallel Neuromorphic Cortex Simulator.

This paper presents a massively parallel and scalable neuromorphic cortex simulator designed for simulating large and structurally connected spiking neural networks, such as complex models of various areas of the cortex. The main novelty of this work is the abstraction of a neuromorphic architecture into clusters represented by minicolumns and hypercolumns, analogously to the fundamental structural units observed in neurobiology. Without this approach, simulating large-scale fully connected networks needs prohibitively large memory to store look-up tables for point-to-point connections. Instead, we use a novel architecture, based on the structural connectivity in the neocortex, such that all the required parameters and connections can be stored in on-chip memory. The cortex simulator can be easily reconfigured for simulating different neural networks without any change in hardware structure by programming the memory. A hierarchical communication scheme allows one neuron to have a fan-out of up to 200 k neurons. As a proof-of-concept, an implementation on one Altera Stratix V FPGA was able to simulate 20 million to 2.6 billion leaky-integrate-and-fire (LIF) neurons in real time. We verified the system by emulating a simplified auditory cortex (with 100 million neurons). This cortex simulator achieved a low power dissipation of 1.62 µW per neuron. With the advent of commercially available FPGA boards, our system offers an accessible and scalable tool for the design, real-time simulation, and analysis of large-scale spiking neural networks.

A pneumatic Bionic Voice prosthesis-Pre-clinical trials of controlling the voice onset and offset.

Despite emergent progress in many fields of bionics, a functional Bionic Voice prosthesis for laryngectomy patients (larynx amputees) has not yet been achieved, leading to a lifetime of vocal disability for these patients. This study introduces a novel framework of Pneumatic Bionic Voice Prostheses as an electronic adaptation of the Pneumatic Artificial Larynx (PAL) device. The PAL is a non-invasive mechanical voice source, driven exclusively by respiration with an exceptionally high voice quality, comparable to the existing gold standard of Tracheoesophageal (TE) voice prosthesis. Following PAL design closely as the reference, Pneumatic Bionic Voice Prostheses seem to have a strong potential to substitute the existing gold standard by generating a similar voice quality while remaining non-invasive and non-surgical. This paper designs the first Pneumatic Bionic Voice prosthesis and evaluates its onset and offset control against the PAL device through pre-clinical trials on one laryngectomy patient. The evaluation on a database of more than five hours of continuous/isolated speech recordings shows a close match between the onset/offset control of the Pneumatic Bionic Voice and the PAL with an accuracy of 98.45 ±0.54%. When implemented in real-time, the Pneumatic Bionic Voice prosthesis controller has an average onset/offset delay of 10 milliseconds compared to the PAL. Hence it addresses a major disadvantage of previous electronic voice prostheses, including myoelectric Bionic Voice, in meeting the short time-frames of controlling the onset/offset of the voice in continuous speech.