Artificial Spiking Neural Networks Research Articles

A Scalable Artificial Neuron Based on Ultrathin Two-Dimensional Titanium Oxide

A spiking neural network consists of artificial synapses and neurons and may realize human-level intelligence. Unlike the widely reported artificial synapses, the fabrication of large-scale artificial neurons with good performance is still challenging due to the lack of a suitable material system and integration method. Here, we report an ultrathin (less than10 nm) and inch-size two-dimensional (2D) oxide-based artificial neuron system produced by a controllable assembly of solution-processed 2D monolayer TiOx nanosheets. Artificial neuron devices based on such 2D TiOx films show a high on/off ratio of 109 and a volatile resistance switching phenomenon. The devices can not only emulate the leaky integrate-and-fire activity but also self-recover without additional circuits for sensing and reset. Moreover, the artificial neuron arrays are fabricated and exhibited good uniformity, indicating their large-area integration potential. Our results offer a strategy for fabricating large-scale and ultrathin 2D material-based artificial neurons and 2D spiking neural networks.

Role of Noise in Spontaneous Activity of Networks of Neurons on Patterned Silicon Emulated by Noise–activated CMOS Neural Nanoelectronic Circuits

Background noise in biological cortical microcircuits constitutes a powerful resource to assess their computational tasks, including, for instance, the synchronization of spiking activity, the enhancement of the speed of information transmission, and the minimization of the corruption of signals. We explore the correlation of spontaneous firing activity of ≈ 100 biological neurons adhering to engineered scaffolds by governing the number of functionalized patterned connection pathways among groups of neurons. We then emulate the biological system by a series of noise-activated silicon neural network simulations. We show that by suitably tuning both the amplitude of noise and the number of synapses between the silicon neurons, the same controlled correlation of the biological population is achieved. Our results extend to a realistic silicon nanoelectronics neuron design using noise injection to be exploited in artificial spiking neural networks such as liquid state machines and recurrent neural networks for stochastic computation.

An artificial spiking quantum neuron

Artificial spiking neural networks have found applications in areas where the temporal nature of activation offers an advantage, such as time series prediction and signal processing. To improve their efficiency, spiking architectures often run on custom-designed neuromorphic hardware, but, despite their attractive properties, these implementations have been limited to digital systems. We describe an artificial quantum spiking neuron that relies on the dynamical evolution of two easy to implement Hamiltonians and subsequent local measurements. The architecture allows exploiting complex amplitudes and back-action from measurements to influence the input. This approach to learning protocols is advantageous in the case where the input and output of the system are both quantum states. We demonstrate this through the classification of Bell pairs which can be seen as a certification protocol. Stacking the introduced elementary building blocks into larger networks combines the spatiotemporal features of a spiking neural network with the non-local quantum correlations across the graph.

A Performance Comparison of Artificial Neural Networks and Spiking Neural Networks

The continual development of artificial neural networks (ANNs) on von-Neuman based hardware poses a threat to potential advances in artificial intelligence (AI). At current rates, power usage of such models will rival that of large cities or even supersede the amount of power produced globally. Therefore, it is imperative to make a transition to a more power efficient computing architecture to make way for future advances in AI. The development of neuromorphic hardware has paved a path for a new set of neural networks called spiking neural networks (SNNs) to act as a viable alternative for machine learning tasks on such brain-like hardware. The primary motive of the research is to contribute towards a search for viable applications of SNNs in their current infancy. The project wishes to address whether SNNs can outperform ANNs image classification of handwritten digits on the MNIST dataset. Despite training on non-optimal von Neumann hardware, statistically significant evidence suggests that the SNN outperformed the ANN in terms of the mean accuracy of the task. As a result of the findings, the paper suggests that further research should focus on translating older ANN models to SNN counterparts that outperform on their tasks and finding applications of SNNs on von Neumann hardware that outperform ANNs. Both prior focuses will narrow the gap between ANN and SNN performance, and in doing so makes a transition to more power efficient and uninhibited AI advancement a reality.

Sparse coding with a somato-dendritic rule

Cortical neurons are silent most of the time: sparse activity enables low-energy computation in the brain, and promises to do the same in neuromorphic hardware. Beyond power efficiency, sparse codes have favourable properties for associative learning, as they can store more information than local codes but are easier to read out than dense codes. Auto-encoders with a sparse constraint can learn sparse codes, and so can single-layer networks that combine recurrent inhibition with unsupervised Hebbian learning. But the latter usually require fast homeostatic plasticity, which could lead to catastrophic forgetting in embodied agents that learn continuously. Here we set out to explore whether plasticity at recurrent inhibitory synapses could take up that role instead, regulating both the population sparseness and the firing rates of individual neurons. We put the idea to the test in a network that employs compartmentalised inputs to solve the task: rate-based dendritic compartments integrate the feedforward input, while spiking integrate-and-fire somas compete through recurrent inhibition. A somato-dendritic learning rule allows somatic inhibition to modulate nonlinear Hebbian learning in the dendrites. Trained on MNIST digits and natural images, the network discovers independent components that form a sparse encoding of the input and support linear decoding. These findings confirm that intrinsic homeostatic plasticity is not strictly required for regulating sparseness: inhibitory synaptic plasticity can have the same effect. Our work illustrates the usefulness of compartmentalised inputs, and makes the case for moving beyond point neuron models in artificial spiking neural networks.

Dual exploration strategies using artificial spiking neural networks in a robotic learning task

Spatial information can be valuable, but new environments may be perceived as risky and thus often evoke fear responses and risk-averse exploration strategies such as thigmotaxis or wall-following behavior. Individual differences in risk-taking (boldness) and thigmotaxis have been reported in natural taxa, which may benefit their survival. In neurorobotic, the common approach is to reproduce cognitive phenomena with multiple levels of bio-inspiration into robotic scenarios. Since autonomous robots may benefit from these different behaviors in exploration tasks, this study aims at simulating two exploration strategies in a virtual robot controlled by a spiking neural network. The experimental context consists in a visual learning task solved through an operant conditioning procedure. Results suggest that the proposed neural architecture sustains both behaviors, switching from one to the other by external cues. This original bio-inspired model could be used as a first step toward further investigations of neurorobotic personality modulated by learning and complex exploration contexts.

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.

An Attention-Based Spiking Neural Network for Unsupervised Spike-Sorting.

Bio-inspired computing using artificial spiking neural networks promises performances outperforming currently available computational approaches. Yet, the number of applications of such networks remains limited due to the absence of generic training procedures for complex pattern recognition, which require the design of dedicated architectures for each situation. We developed a spike-timing-dependent plasticity (STDP) spiking neural network (SSN) to address spike-sorting, a central pattern recognition problem in neuroscience. This network is designed to process an extracellular neural signal in an online and unsupervised fashion. The signal stream is continuously fed to the network and processed through several layers to output spike trains matching the truth after a short learning period requiring only few data. The network features an attention mechanism to handle the scarcity of action potential occurrences in the signal, and a threshold adaptation mechanism to handle patterns with different sizes. This method outperforms two existing spike-sorting algorithms at low signal-to-noise ratio (SNR) and can be adapted to process several channels simultaneously in the case of tetrode recordings. Such attention-based STDP network applied to spike-sorting opens perspectives to embed neuromorphic processing of neural data in future brain implants.

A CMOS Compatible Bulk FinFET-Based Ultra Low Energy Leaky Integrate and Fire Neuron for Spiking Neural Networks

The fundamental building block of an artificial spiking neural network (SNN) is an element which can effectively mimic a biological neuron. There are several electronic and spintronic devices which have been demonstrated as a neuron. But the main concern here is the energy consumption and large area of those artificial neurons. In this letter, we propose and demonstrate a highly scalable and CMOS compatible bulk FinFET with an ${n}^{+}$ buried layer for ultra low energy artificial neuron using well calibrated TCAD simulations. The proposed device shows the signature spiking frequency versus input voltage curve of a biological neuron. The energy per spike of the integrate block of the proposed leaky integrate and fire (LIF) neuron is ${6.3}~\textit {fJ}$ /spike which is the minimum reported till date.

Artificial Neuron and Synapse Realized in an Antiferromagnet/Ferromagnet Heterostructure Using Dynamics of Spin-Orbit Torque Switching.

Efficient information processing in the human brain is achieved by dynamics of neurons and synapses, motivating effective implementation of artificial spiking neural networks. Here, the dynamics of spin-orbit torque switching in antiferromagnet/ferromagnet heterostructures is studied to show the capability of the material system to form artificial neurons and synapses for asynchronous spiking neural networks. The magnetization switching, driven by a single current pulse or trains of pulses, is examined as a function of the pulse width (1 s to 1 ns), amplitude, number, and pulse-to-pulse interval. Based on this dynamics and the unique ability of the system to exhibit binary or analog behavior depending on the device size, key functionalities of a synapse (spike-timing-dependent plasticity) and a neuron (leaky integrate-and-fire) are reproduced in the same material and on the basis of the same working principle. These results open a way toward spintronics-based neuromorphic hardware that executes cognitive tasks with the efficiency of the human brain.

Spatial Concept Learning: A Spiking Neural Network Implementation in Virtual and Physical Robots.

This paper proposes an artificial spiking neural network (SNN) sustaining the cognitive abstract process of spatial concept learning, embedded in virtual and real robots. Based on an operant conditioning procedure, the robots learn the relationship of horizontal/vertical and left/right visual stimuli, regardless of their specific pattern composition or their location on the images. Tests with novel patterns and locations were successfully completed after the acquisition learning phase. Results show that the SNN can adapt its behavior in real time when the rewarding rule changes.

Spiking Neurons Integrating Visual Stimuli Orientation and Direction Selectivity in a Robotic Context.

Visual motion detection is essential for the survival of many species. The phenomenon includes several spatial properties, not fully understood at the level of a neural circuit. This paper proposes a computational model of a visual motion detector that integrates direction and orientation selectivity features. A recent experiment in the Drosophila model highlights that stimulus orientation influences the neural response of direction cells. However, this interaction and the significance at the behavioral level are currently unknown. As such, another objective of this article is to study the effect of merging these two visual processes when contextualized in a neuro-robotic model and an operant conditioning procedure. In this work, the learning task was solved using an artificial spiking neural network, acting as the brain controller for virtual and physical robots, showing a behavior modulation from the integration of both visual processes.

Passive memristor synaptic circuits with multiple timing dependent plasticity mechanisms

Adaptation of synaptic strength is central to memory and learning in biological systems, enabling important high-level processes such as the ability of animals to respond to a changing environment. Memristor devices are a promising new, nanoscale technology that emulates the function of synapses and can therefore be used to represent synaptic elements in analog artificial neural networks. The main mechanism to carry out unsupervised adaptation of weights in memristive synapses currently involves artificial spiking neural network designs relying on spike-timing dependent plasticity (STDP). We present and analyze a new memristive circuit that in addition to STDP learning rules also implements competitive adjustment based on relative timing of presynaptic inputs. The cooperative effect of multiple learning rules in the new circuit may ameliorate the problem of erasure of synaptic weights.

Role of synaptic variability in resistive memory-based spiking neural networks with unsupervised learning

Resistive switching memories (RRAMs) have attracted wide interest as adaptive synaptic elements in artificial bio-inspired spiking neural networks (SNNs). These devices suffer from high cycle-to-cycle and cell-to-cell conductance variability, which is usually considered as a big challenge. However, biological synapses are noisy devices and the brain seems in some situations to benefit from the noise. It has been predicted that RRAM-based SNNs are intrinsically robust to synaptic variability. Here, we investigate this robustness based on extensive characterization data: we analyze the role of noise during unsupervised learning by spike-timing dependent plasticity (STDP) for detection in dynamic input data and classification of static input data. Extensive characterizations of multi-kilobits HfO2-based oxide-based RAM (OxRAM) arrays under different programming conditions are presented. We identify the trade-offs between programming conditions, power consumption, conductance variability and endurance features. Finally, the experimental results are used to perform system-level simulations fully calibrated on the experimental data. The results demonstrate that, similarly to biology, SNNs are not only robust to noise but a certain amount of noise can even improve the network performance. OxRAM conductance variability increases the range of synaptic values explored during the learning process. Moreover, the reduction of constraints on the OxRAM conductance variability allows the system to operate at low power programming conditions.

Learning Nonlinear Dynamics in Efficient, Balanced Spiking Networks Using Local Plasticity Rules

The brain uses spikes in neural circuits to perform many dynamical computations. The computations are performed with properties such as spiking efficiency, i.e. minimal number of spikes, and robustness to noise. A major obstacle for learning computations in artificial spiking neural networks with such desired biological properties is due to lack of our understanding of how biological spiking neural networks learn computations. Here, we consider the credit assignment problem, i.e. determining the local contribution of each synapse to the network's global output error, for learning nonlinear dynamical computations in a spiking network with the desired properties of biological networks. We approach this problem by fusing the theory of efficient, balanced neural networks (EBN) with nonlinear adaptive control theory to propose a local learning rule. Locality of learning rules are ensured by feeding back into the network its own error, resulting in a learning rule depending solely on presynaptic inputs and error feedbacks. The spiking efficiency and robustness of the network are guaranteed by maintaining a tight excitatory/inhibitory balance, ensuring that each spike represents a local projection of the global output error and minimizes a loss function. The resulting networks can learn to implement complex dynamics with very small numbers of neurons and spikes, exhibit the same spike train variability as observed experimentally, and are extremely robust to noise and neuronal loss.

Sameness/difference spiking neural circuit as a relational concept precursor model: A bio-inspired robotic implementation

Abstract This paper presents a bio-inspired virtual and physical robotic model of sameness/difference (SD) abstract relations investigated through associative learning tasks. Considering that the invertebrate bees presumably possess this higher cognitive ability with a relatively tiny brain (Giurfa, Zhang, Jenett, Menzel, & Srinivasan, 2001), an hypothesis of a neural correlate may hold as a minimalist cellular circuit. In order to simulate it, we implemented the model using an artificial spiking neural network (SNN) acting as a robot’s brain-controller and a reward-modulated spike-timing dependent plasticity as the learning rule. The model is tested with different operant conditioning procedures of delayed-matching-to-sample and delayed-non-matching-to-sample tasks from dual choices of colors. Apart SD, the color and side simpler associative rules also co-exist. This allows the robot to learn in different scenarios, not knowing the rule to come. Therefore, we showed that depending on its action and the applied reinforcing rules (SD, color or side), the robot dynamically learns any of it, independently and one after the other, all supported by a single SNN. We believe that this SD SNN model could be used as a precursor base to understand and build a generic form of a relational concept mechanism.

Spiking Neural Networks Based on OxRAM Synapses for Real-Time Unsupervised Spike Sorting.

In this paper, we present an alternative approach to perform spike sorting of complex brain signals based on spiking neural networks (SNN). The proposed architecture is suitable for hardware implementation by using resistive random access memory (RRAM) technology for the implementation of synapses whose low latency (<1μs) enables real-time spike sorting. This offers promising advantages to conventional spike sorting techniques for brain-computer interfaces (BCI) and neural prosthesis applications. Moreover, the ultra-low power consumption of the RRAM synapses of the spiking neural network (nW range) may enable the design of autonomous implantable devices for rehabilitation purposes. We demonstrate an original methodology to use Oxide based RRAM (OxRAM) as easy to program and low energy (<75 pJ) synapses. Synaptic weights are modulated through the application of an online learning strategy inspired by biological Spike Timing Dependent Plasticity. Real spiking data have been recorded both intra- and extracellularly from an in-vitro preparation of the Crayfish sensory-motor system and used for validation of the proposed OxRAM based SNN. This artificial SNN is able to identify, learn, recognize and distinguish between different spike shapes in the input signal with a recognition rate about 90% without any supervision.

Leaky Integrate-and-Fire Neuron Circuit Based on Floating-Gate Integrator.

The artificial spiking neural network (SNN) is promising and has been brought to the notice of the theoretical neuroscience and neuromorphic engineering research communities. In this light, we propose a new type of artificial spiking neuron based on leaky integrate-and-fire (LIF) behavior. A distinctive feature of the proposed FG-LIF neuron is the use of a floating-gate (FG) integrator rather than a capacitor-based one. The relaxation time of the charge on the FG relies mainly on the tunnel barrier profile, e.g., barrier height and thickness (rather than the area). This opens up the possibility of large-scale integration of neurons. The circuit simulation results offered biologically plausible spiking activity (<100 Hz) with a capacitor of merely 6 fF, which is hosted in an FG metal-oxide-semiconductor field-effect transistor. The FG-LIF neuron also has the advantage of low operation power (<30 pW/spike). Finally, the proposed circuit was subject to possible types of noise, e.g., thermal noise and burst noise. The simulation results indicated remarkable distributional features of interspike intervals that are fitted to Gamma distribution functions, similar to biological neurons in the neocortex.

Spiking Neural Network With Distributed Plasticity Reproduces Cerebellar Learning in Eye Blink Conditioning Paradigms.

In this study, we defined a realistic cerebellar model through the use of artificial spiking neural networks, testing it in computational simulations that reproduce associative motor tasks in multiple sessions of acquisition and extinction. By evolutionary algorithms, we tuned the cerebellar microcircuit to find out the near-optimal plasticity mechanism parameters that better reproduced human-like behavior in eye blink classical conditioning, one of the most extensively studied paradigms related to the cerebellum. We used two models: one with only the cortical plasticity and another including two additional plasticity sites at nuclear level. First, both spiking cerebellar models were able to well reproduce the real human behaviors, in terms of both "timing" and "amplitude", expressing rapid acquisition, stable late acquisition, rapid extinction, and faster reacquisition of an associative motor task. Even though the model with only the cortical plasticity site showed good learning capabilities, the model with distributed plasticity produced faster and more stable acquisition of conditioned responses in the reacquisition phase. This behavior is explained by the effect of the nuclear plasticities, which have slow dynamics and can express memory consolidation and saving. We showed how the spiking dynamics of multiple interactive neural mechanisms implicitly drive multiple essential components of complex learning processes. This study presents a very advanced computational model, developed together by biomedical engineers, computer scientists, and neuroscientists. Since its realistic features, the proposed model can provide confirmations and suggestions about neurophysiological and pathological hypotheses and can be used in challenging clinical applications.

Wavelet decomposition and phase encoding of temporal signals using spiking neurons

Spike encoding is the initial yet crucial step for any application domain of Artificial Spiking Neural Networks (ASNN). However, current encoding methods are not suitable to process complex temporal signal. Motivated by the modulation relationship found between afferent synaptic currents in biological neurons, this study proposes a spike phase encoding method for ASNN, which could perform wavelet decomposition on the input signal, and encode the wavelet spectrum into synchronized output spike trains. The spike delays in each synchronizing period represent the spectrum amplitudes. The encoding method was tested in two implementation examples: (a) encoding of human voice records for speech recognition propose; and (b) encoding of multichannel electroencephalography (EEG) records with the aim to detect interictal spikes in patients with epilepsy. Empirical evaluations confirm that encoded spike trains constitute a good representation of the continuous wavelet transform of the original signal, with the ability to capture interesting features from the input signal.