Spiking Neural Network Model Research Articles

Prediction of tinnitus masking benefit within a case series using a spiking neural network model.

Masking has been widely used as a tinnitus therapy, with large individual differences in its effectiveness. The basis of this variation is unknown. We examined individual tinnitus and psychological responses to three masking types, energetic masking (bilateral broadband static or rain noise [BBN]), informational masking (BBN with a notch at tinnitus pitch and 3-dimensional cues) and a masker combining both effects (BBN with spatial cues). Eleven participants with chronic tinnitus were followed for 12 months, each person used each masking approach for 3 months with a 1 month washout-baseline. The Tinnitus Functional Index (TFI), Tinnitus Rating Scales, Positive and Negative Affect Scale and Depression Anxiety Stress Scales, were measured every month of treatment. Electroencephalography (EEG) and psychoacoustic assessment was undertaken at baseline and following 3 months of each masking sound. The computational modeling of EEG data was based on the framework of brain-inspired Spiking Neural Network (SNN) architecture called NeuCube, designed for this study for mapping, learning, visualizing and classifying of brain activity patterns. EEG was related to clinically significant change in the TFI using the SNN model. The SNN framework was able to predict sound therapy responders (93% accuracy) from non-responders (100% accuracy) using baseline EEG recordings. The combination of energetic and informational masking was an effective treatment sound in more individuals than the other sounds used. Although the findings are promising, they are preliminary and require confirmation in independent and larger samples.

Optimization of Spiking Neural Networks Based on Binary Streamed Rate Coding

Spiking neural networks (SNN) increasingly attract attention for their similarity to the biological neural system. Hardware implementation of spiking neural networks, however, remains a great challenge due to their excessive complexity and circuit size. This work introduces a novel optimization method for hardware friendly SNN architecture based on a modified rate coding scheme called Binary Streamed Rate Coding (BSRC). BSRC combines the features of both rate and temporal coding. In addition, by employing a built-in randomizer, the BSRC SNN model provides a higher accuracy and faster training. We also present SNN optimization methods including structure optimization and weight quantization. Extensive evaluations with MNIST SNNs demonstrate that the structure optimization of SNN (81-30-20-10) provides 183.19 times reduction in hardware compared with SNN (784-800-10), while providing an accuracy of 95.25%, a small loss compared with 98.89% and 98.93% reported in the previous works. Our weight quantization reduces 32-bit weights to 4-bit integers leading to further hardware reduction of 4 times with only 0.56% accuracy loss. Overall, the SNN model (81-30-20-10) optimized by our method shrinks the SNN’s circuit area from 3089.49 mm2 for SNN (784-800-10) to 4.04 mm2—a reduction of 765 times.

Enabling Resource-Aware Mapping of Spiking Neural Networks via Spatial Decomposition

With growing model complexity, mapping spiking neural network (SNN)-based applications to tile-based neuromorphic hardware is becoming increasingly challenging. This is because the synaptic storage resources on a tile, viz., a crossbar, can accommodate only a fixed number of presynaptic connections per postsynaptic neuron. For complex SNN models that have many presynaptic connections per neuron, some connections may need to be pruned after training to fit onto the tile resources, leading to a loss in the model quality, e.g., accuracy. In this letter, we propose a novel unrolling technique that decomposes a neuron function with many presynaptic connections into a sequence of homogeneous neural units, where each neural unit is a function computation node, with two presynaptic connections. This spatial decomposition technique significantly improves crossbar utilization and retains all presynaptic connections, resulting in no loss of the model quality derived from connection pruning. We integrate the proposed technique within an existing SNN mapping framework and evaluate it using machine learning applications on the DYNAP-SE state-of-the-art neuromorphic hardware. Our results demonstrate an average 60% lower crossbar requirement, 9× higher synapse utilization, 62% lower wasted energy on the hardware, and between 0.8% and 4.6% increase in the model quality.

A three-tier road condition classification system using a spiking neural network model

Road surface anomaly detection and classification based on crowd-sourced smart phone sensor data has become an important area of research over the last decade due to its potential benefits to road maintenance. Previous studies focused on paved roads in which anomaly classification were modelled as single-staged events mostly using machine learning and threshold-based methods. Little or no attention has been paid to road type classification and anomaly detection and classification on unpaved roads, which constitute a larger percentage of roads in the developing world. In this paper, road condition classification is approached as a multi-tier activity, comprising of models for road type classification, anomaly classification models for paved roads as well as unpaved roads using a novel Spiking Neural Network (SNN) learning model. To demonstrate the viability of the proposed system, road condition data for the various tasks were collected via an Android Application developed by the authors from which statistical features were extracted and used to train and evaluate the models. Experimental results showed that the proposed SNN model yielded significantly higher classification performance when compared to a Support Vector Machine (SVM) and Multilayer Perceptron (MLP) trained and tested using the collected datasets and classification models reported in existing studies.

A Model of Memory Linking Time to Space.

The storage of temporally precise spike patterns can be realized by a single neuron. A spiking neural network (SNN) model is utilized to demonstrate the ability to precisely recall a spike pattern after presenting a single input. We show by using a simulation study that the temporal properties of input patterns can be transformed into spatial patterns of local dendritic spikes. The localization of time-points of spikes is facilitated by phase-shift of the subthreshold membrane potential oscillations (SMO) in the dendritic branches, which modifies their excitability. In reference to the points in time of the arriving input, the dendritic spikes are triggered in different branches. To store spatially distributed patterns, two unsupervised learning mechanisms are utilized. Either synaptic weights to the branches, spatial representation of the temporal input pattern, are enhanced by spike-timing-dependent plasticity (STDP) or the oscillation power of SMOs in spiking branches is increased by dendritic spikes. For retrieval, spike bursts activate stored spatiotemporal patterns in dendritic branches, which reactivate the original somatic spike patterns. The simulation of the prototypical model demonstrates the principle, how linking time to space enables the storage of temporal features of an input. Plausibility, advantages, and some variations of the proposed model are also discussed.

A biologically constrained spiking neural network model of the primate basal ganglia with overlapping pathways exhibits action selection.

Action selection has been hypothesized to be a key function of the basal ganglia, yet the nuclei involved, their interactions and the importance of the direct/indirect pathway segregation in such process remain debated. Here, we design a spiking computational model of the monkey basal ganglia derived from a previously published population model, initially parameterized to reproduce electrophysiological activity at rest and to embody as much quantitative anatomical data as possible. As a particular feature, both models exhibit the strong overlap between the direct and indirect pathways that has been documented in non‐human primates. Here, we first show how the translation from a population to an individual neuron model was achieved, with the addition of a minimal number of parameters. We then show that our model performs action selection, even though it was built without any assumption on the activity carried out during behaviour. We investigate the mechanisms of this selection through circuit disruptions and found an instrumental role of the off‐centre/on‐surround structure of the MSN‐STN‐GPi circuit, as well as of the MSN‐MSN and FSI‐MSN projections. This validates their potency in enabling selection. We finally study the pervasive centromedian and parafascicular thalamic inputs that reach all basal ganglia nuclei and whose influence is therefore difficult to anticipate. Our model predicts that these inputs modulate the responsiveness of action selection, making them a candidate for the regulation of the speed–accuracy trade‐off during decision‐making.

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.

Correction: Correction: The role of cortical oscillations in a spiking neural network model of the basal ganglia.

[This corrects the article DOI: 10.1371/journal.pone.0205472.].

Probabilistic inference of binary Markov random fields in spiking neural networks through mean-field approximation

Recent studies have suggested that the cognitive process of the human brain is realized as probabilistic inference and can be further modeled by probabilistic graphical models like Markov random fields. Nevertheless, it remains unclear how probabilistic inference can be implemented by a network of spiking neurons in the brain. Previous studies have tried to relate the inference equation of binary Markov random fields to the dynamic equation of spiking neural networks through belief propagation algorithm and reparameterization, but they are valid only for Markov random fields with limited network structure. In this paper, we propose a spiking neural network model that can implement inference of arbitrary binary Markov random fields. Specifically, we design a spiking recurrent neural network and prove that its neuronal dynamics are mathematically equivalent to the inference process of Markov random fields by adopting mean-field theory. Furthermore, our mean-field approach unifies previous works. Theoretical analysis and experimental results, together with the application to image denoising, demonstrate that our proposed spiking neural network can get comparable results to that of mean-field inference.

An Indexing Theory for Working Memory Based on Fast Hebbian Plasticity.

Working memory (WM) is a key component of human memory and cognition. Computational models have been used to study the underlying neural mechanisms, but neglected the important role of short-term memory (STM) and long-term memory (LTM) interactions for WM. Here, we investigate these using a novel multiarea spiking neural network model of prefrontal cortex (PFC) and two parietotemporal cortical areas based on macaque data. We propose a WM indexing theory that explains how PFC could associate, maintain, and update multimodal LTM representations. Our simulations demonstrate how simultaneous, brief multimodal memory cues could build a temporary joint memory representation as an “index” in PFC by means of fast Hebbian synaptic plasticity. This index can then reactivate spontaneously and thereby also the associated LTM representations. Cueing one LTM item rapidly pattern completes the associated uncued item via PFC. The PFC–STM network updates flexibly as new stimuli arrive, thereby gradually overwriting older representations.

Pre-Synaptic Pool Modification (PSPM): A supervised learning procedure for recurrent spiking neural networks.

Learning synaptic weights of spiking neural network (SNN) models that can reproduce target spike trains from provided neural firing data is a central problem in computational neuroscience and spike-based computing. The discovery of the optimal weight values can be posed as a supervised learning task wherein the weights of the model network are chosen to maximize the similarity between the target spike trains and the model outputs. It is still largely unknown whether optimizing spike train similarity of highly recurrent SNNs produces weight matrices similar to those of the ground truth model. To this end, we propose flexible heuristic supervised learning rules, termed Pre-Synaptic Pool Modification (PSPM), that rely on stochastic weight updates in order to produce spikes within a short window of the desired times and eliminate spikes outside of this window. PSPM improves spike train similarity for all-to-all SNNs and makes no assumption about the post-synaptic potential of the neurons or the structure of the network since no gradients are required. We test whether optimizing for spike train similarity entails the discovery of accurate weights and explore the relative contributions of local and homeostatic weight updates. Although PSPM improves similarity between spike trains, the learned weights often differ from the weights of the ground truth model, implying that connectome inference from spike data may require additional constraints on connectivity statistics. We also find that spike train similarity is sensitive to local updates, but other measures of network activity such as avalanche distributions, can be learned through synaptic homeostasis.

An end-to-end functional spiking model for sequential feature learning

Spiking Neural Network (SNN) has recently gained significant momentum in the neuromorphic low-power systems. However, the existing SNN models have limited use in time-sequential feature learning, and the exhausting spike encoding and decoding make the SNNs not straightforward to use. Inspired by the functional organization in the primate visual system, we propose an end-to-end functional spiking model in this paper to address these issues. Specifically, we propose the functional spike response to make each neuron special, and the dynamic synaptic efficiency to make the transmission of each input signal controllable. We represent inputs by a simple two-tuple set instead of conventional complex encoding, which achieves end-to-end learning. Experiments on synthetic datasets demonstrate that employing the two-tuple encoding strategy, our method improves the accuracy of the traditional SNN model significantly. In addition, we apply our method to seven real-world datasets and one human motion prediction dataset to investigate its performance. Experimental results show that the proposed functional spike response organization saves the running time of our model compared with the LSTM, GRU and one of the state-of-the-art time series processing methods.

A Curiosity-Based Learning Method for Spiking Neural Networks

Spiking Neural Networks (SNNs) have shown favorable performance recently. Nonetheless, the time-consuming computation on neuron level and complex optimization limit their real-time application. Curiosity has shown great performance in brain learning, which helps biological brains grasp new knowledge efficiently and actively. Inspired by this leaning mechanism, we propose a curiosity-based SNN (CBSNN) model, which contains four main learning processes. Firstly, the network is trained with biologically plausible plasticity principles to get the novelty estimations of all samples in only one epoch; secondly, the CBSNN begins to repeatedly learn the samples whose novelty estimations exceed the novelty threshold and dynamically update the novelty estimations of samples according to the learning results in five epochs; thirdly, in order to avoid the overfitting of the novel samples and forgetting of the learned samples, CBSNN retrains all samples in one epoch; finally, step two and step three are periodically taken until network convergence. Compared with the state-of-the-art Voltage-driven Plasticity-centric SNN (VPSNN) under standard architecture, our model achieves a higher accuracy of 98.55% with only 54.95% of its computation cost on the MNIST hand-written digit recognition dataset. Similar conclusion can also be found out in other datasets, i.e., Iris, NETtalk, Fashion-MNIST, and CIFAR-10, respectively. More experiments and analysis further prove that such curiosity-based learning theory is helpful in improving the efficiency of SNNs. As far as we know, this is the first practical combination of the curiosity mechanism and SNN, and these improvements will make the realistic application of SNNs possible on more specific tasks within the von Neumann framework.

Low-Energy and Fast Spiking Neural Network For Context-Dependent Learning on FPGA

Supervised, unsupervised, and reinforcement learning (RL) mechanisms are known as the most powerful learning paradigms empowering neuromorphic systems. These systems typically take advantage of unsupervised learning because they can learn the distribution of sensory information. However, to perform a task, not only is it important to have sensory information, but also it is required to have information about the context in which the system is operating. In this sense, reinforcement learning is very powerful for interacting with the environment while performing a context-dependent task. The predominant motivation for this brief is to present a digital architecture for a spiking neural network (SNN) model with RL capability suitable for learning a context-dependent task. The proposed architecture is composed of hardware-friendly leaky integrate-and-firing (LIF) neurons and spike timing dependent plasticity (STDP)-based synapses implemented on a field programmable gate array (FPGA). Hardware synthesis and physical implementations show that the resulting circuits can faithfully reproduce the outcome of a learning task previously performed in both animal experimentation and computational modelings. Compared to the state-of-the-art neuromorphic FPGA circuits with context-dependent learning capability, our circuit fires 10.7 times fewer spikes, which accelerates learning 15 times, while requiring 16 times less energy. This is a significant step in achieving fast and low-energy SNNs with context-dependent learning ability on FPGAs.

Hardware and Software Optimizations for Accelerating Deep Neural Networks: Survey of Current Trends, Challenges, and the Road Ahead

Currently, Machine Learning (ML) is becoming ubiquitous in everyday life. Deep Learning (DL) is already present in many applications ranging from computer vision for medicine to autonomous driving of modern cars as well as other sectors in security, healthcare, and finance. However, to achieve impressive performance, these algorithms employ very deep networks, requiring a significant computational power, both during the training and inference time. A single inference of a DL model may require billions of multiply-and-accumulated operations, making the DL extremely compute- and energy-hungry. In a scenario where several sophisticated algorithms need to be executed with limited energy and low latency, the need for cost-effective hardware platforms capable of implementing energy-efficient DL execution arises. This paper first introduces the key properties of two brain-inspired models like Deep Neural Network (DNN), and Spiking Neural Network (SNN), and then analyzes techniques to produce efficient and high-performance designs. This work summarizes and compares the works for four leading platforms for the execution of algorithms such as CPU, GPU, FPGA and ASIC describing the main solutions of the state-of-the-art, giving much prominence to the last two solutions since they offer greater design flexibility and bear the potential of high energy-efficiency, especially for the inference process. In addition to hardware solutions, this paper discusses some of the important security issues that these DNN and SNN models may have during their execution, and offers a comprehensive section on benchmarking, explaining how to assess the quality of different networks and hardware systems designed for them.

A Multi-Spiking Neural Network Learning Model for Data Classification

Classical Artificial Neural Networks (ANNs) though well exploited in solving classification problems, do not model perfectly the information encoding process in the human brain because ANNs encode information using rate-based coding. However, biological neurons in the brain are known to encode information using temporal coding. In order to mimic the biological method of encoding information, various Spiking Neural Network (SNN) models have been developed. However, some of these models are limited in the number of spikes and do not leverage well on some classification problems. In order to address some of the inherent challenges associated with SNN, a multi-layer learning model for a multi-spiking network is proposed in this paper. The model exploits the temporal coding of spikes and the least-squares method to derive a weight update scheme. It also employs a spike locality concept in order to determine how the synaptic weights are to be adjusted at a particular spike time so as to minimize the learning interference, and thereby, increasing the number of spikes for learning. The performance of the model is evaluated on benchmarked classification datasets. A correlation-based metric is combined with a threshold concept to measure the classification accuracy of the model. The experimental results showed that the proposed model achieved better classification accuracy than some state-of-the-art multi-layer SNN learning models.

A spiking neural network model of spatial and visual mental imagery.

Mental imagery has long been of interest to the cognitive and neurosciences, but how it manifests itself in the mind and brain still remains unresolved. In pursuit of this, we built a spiking neural model that can perform mental rotation and mental map scanning using strategies informed by the psychology and neuroscience literature. Results: When performing mental map scanning, reaction times (RTs) for our model closely match behavioural studies (approx. 50ms/cm), and replicate the cognitive penetrability of the task. When performing mental rotation, our model's RTs once again closely match behavioural studies (model: 55-65°/s; studies: 60°/s), and performed the task using the same task strategy (whole unit rotation of simple and familiar objects through intermediary points). Overall, our model suggests: (1) vector-based approaches to neuro-cognitive modelling are well equipped to re-produce behavioural findings, and (2) the cognitive (in)penetrability of imagery tasks may depend on whether or not the task makes use of (non)symbolic processing.

Bio-Inspired Evolutionary Model of Spiking Neural Networks in Ionic Liquid Space.

One of the biggest struggles while working with artificial neural networks is being able to come up with models which closely match biological observations. Biological neural networks seem to capable of creating and pruning dendritic spines, leading to synapses being changed, which results in higher learning capability. The latter forms the basis of the present study in which a new ionic model for reservoir-like networks, consisting of spiking neurons, is introduced. High plasticity of this model makes learning possible with a fewer number of neurons. In order to study the effect of the applied stimulus in an ionic liquid space through time, a diffusion operator is used which somehow compensates for the separation between spatial and temporal coding in spiking neural networks and therefore, makes the mentioned model suitable for spatiotemporal patterns. Inspired by partial structural changes in the human brain over the years, the proposed model evolves during the learning process. The effect of topological evolution on the proposed model's performance for some classification problems is studied in this paper. Several datasets have been used to evaluate the performance of the proposed model compared to the original LSM. Classification results via separation and accuracy values have shown that the proposed ionic liquid outperforms the original LSM.

Spiking Neural Networks applied to the classification of motor tasks in EEG signals

Motivated by the recent progress of Spiking Neural Network (SNN) models in pattern recognition, we report on the development and evaluation of brain signal classifiers based on SNNs. The work shows the capabilities of this type of Spiking Neurons in the recognition of motor imagery tasks from EEG signals and compares their performance with other traditional classifiers commonly used in this application. This work includes two stages: the first stage consists of comparing the performance of the SNN models against some traditional neural network models. The second stage, compares the SNN models performance in two input conditions: input features with constant values and input features with temporal information. The EEG signals employed in this work represent five motor imagery tasks: i.e. rest, left hand, right hand, foot and tongue movements. These EEG signals were obtained from a public database provided by the Technological University of Graz (Brunner et al., 2008). The feature extraction stage was performed by applying two algorithms: power spectral density and wavelet decomposition. Likewise, this work uses raw EEG signals for the second stage of the problem solution. All of the models were evaluated in the classification between two motor imagery tasks. This work demonstrates that with a smaller number of Spiking neurons, simple problems can be solved. Better results are obtained by using patterns with temporal information, thereby exploiting the capabilities of the SNNs.

Parameter Optimization and Learning in a Spiking Neural Network for UAV Obstacle Avoidance Targeting Neuromorphic Processors

The Lobula giant movement detector (LGMD) is an identified neuron of the locust that detects looming objects and triggers the insect's escape responses. Understanding the neural principles and network structure that leads to these fast and robust responses can facilitate the design of efficient obstacle avoidance strategies for robotic applications. Here, we present a neuromorphic spiking neural network model of the LGMD driven by the output of a neuromorphic dynamic vision sensor (DVS), which incorporates spiking frequency adaptation and synaptic plasticity mechanisms, and which can be mapped onto existing neuromorphic processor chips. However, as the model has a wide range of parameters and the mixed-signal analog-digital circuits used to implement the model are affected by variability and noise, it is necessary to optimize the parameters to produce robust and reliable responses. Here, we propose to use differential evolution (DE) and Bayesian optimization (BO) techniques to optimize the parameter space and investigate the use of self-adaptive DE (SADE) to ameliorate the difficulties of finding appropriate input parameters for the DE technique. We quantify the performance of the methods proposed with a comprehensive comparison of different optimizers applied to the model and demonstrate the validity of the approach proposed using recordings made from a DVS sensor mounted on an unmanned aerial vehicle (UAV).