Neuronal Network Model Research Articles

Synchronization of Rulkov neuron networks coupled by excitatory and inhibitory chemical synapses.

This paper takes into account a neuron network model in which the excitatory and the inhibitory Rulkov neurons interact each other through excitatory and inhibitory chemical coupling, respectively. Firstly, for two or more identical or non-identical Rulkov neurons, the existence conditions of the synchronization manifold of the fixed points are investigated, which have received less attention over the past decades. Secondly, the master stability equation of the arbitrarily connected neuron network under the existence conditions of the synchronization manifold is discussed. Thirdly, taking three identical Rulkov neurons as an example, some new results are presented: (1) topological structures that can make the synchronization manifold exist are given, (2) the stability of synchronization when different parameters change is discussed, and (3) the roles of the control parameters, the ratio, as well as the size of the coupling strength and sigmoid function are analyzed. Finally, for the chemical coupling between two non-identical neurons, the transversal system is given and the effect of two coupling strengths on synchronization is analyzed.

A leaky integrate and fire model for spike generation in a neuron with variable threshold and multiple‐input–single‐output configuration

AbstractComprehensive understanding of the phenomenon of spike transfer from one neuron to another in human brain is essential for developing spiking neuronal network models. Synapses play pivotal role in transferring information from one neuron to another. A leaky integrate and fire model of a representative neuron in multiple‐input and single‐output configuration with threshold variability for neuronal spike transfer process is proposed in this paper. A simple analytical expression of membrane potential including threshold variability over a small time interval is developed. The model captures several important features of a spiking neuron through a minimal set of model parameters. An activity dependent noise model is also suggested. An expression for the probability of generation of an action potential at axon hillock of the post‐synaptic neuron is derived from the model. Analytical and simulation results are provided to explain various aspects of the proposed model.

Axonal Anatomy Optimizes Spatial Encoding in the Rat Entorhinal-Dentate System: A Computational Study.

The network architecture connecting neural regions is defined by the organization and anatomical properties of the projecting axons, but its contributions to neural encoding and system function are difficult to study experimentally. Using a large-scale, spiking neuronal network model of rat dentate gyrus, the role of the anatomy of the entorhinal-dentate axonal projection was evaluated in the context of spatial encoding by incorporating grid cell activity to provide physiological, spatially-correlated input. The dorso-ventral extents of the entorhinal axon terminal fields were varied to generate different feedforward architectures, and the resulting spatial representations and spatial information scores of the network were evaluated. Position was decoded from the population activity using a point process filter to investigate the contributions of network architecture on spatial encoding. The model predicted the emergence of anatomical gradients within the dentate gyrus for place field size and spatial information along its dorso-ventral axis, which were dependent on the extents of the entorhinal axon terminal fields. The decoding results revealed an optimal performance at an axon terminal field extent of 2mm that lies within the biological range. The axonal anatomy mediates a tradeoff between encoding multiple place field sizes or achieving a high spatial information score, and the combination of both properties is necessary to maximize spatial encoding by a network. In total, this paper establishes a mechanistic neuronal network model that, in concert with information-theoretic and statistical methods, can be used to investigate how lower level properties contribute to higher level function.

Call for community review of PyNN — A simulator-independent language for building neuronal network models

The purpose of this document is to solicit community feedback on PyNN, a simulator-independent language for building neuronal network models, which was submitted to INCF for endorsement as a standard. The document contains the INCF standards and best practices committee's review of PyNN, and the criteria in which it was evaluated (open, FAIR, testing and implementation, governance, adoption and use, stability and support, and comparison to similar standards). For the next 60 days, we are seeking community feedback on PyNN. About PyNN: the PyNN API aims to support modelling at a high-level of abstraction (populations of neurons, layers, columns and the connections between them) while still allowing access to the details of individual neurons and synapses when required. PyNN provides a library of standard neuron, synapse and synaptic plasticity models, which have been verified to work the same on the different supported simulators. PyNN also provides a set of commonly-used connectivity algorithms (e.g. all-to-all, random, distance-dependent, small-world) but makes it easy to provide your own connectivity in a simulator-independent way, either using the Connection Set Algebra or by writing your own Python code.

A radial basis probabilistic process neural network model and corresponding classification algorithm

A radial basis probabilistic process neuron (RBPPN) and radial basis probabilistic process neural network (RBPPNN) model are proposed to fuse a priori knowledge for application to time-varying signal pattern classification. RBPPN inputs were multi-channel time-varying signals and a generalized inner product was used to perform spatio-temporal aggregation of input signals in the kernel. Typical signal samples from various pattern subsets in the sample set were used as kernel center functions, which use morphological distribution characteristics and combination relationships to implicitly express prior knowledge for the signal category. The exponential probability function was used as the activation function to achieve kernel transformation and RBPPN probability output. The RBPPNN is composed of process signal input layers, an RBPPN hidden layer, a pattern layer, and a Softmax classifier developed through stacking. Generalized inner product operations were used to conduct probability similarity measurements of distribution characteristics between process signals. The pattern layer selectivity summed inputs from the RBPPN hidden layer to the pattern layer according to the category of the kernel center function. Its outputs were then used as inputs in the Softmax classifier. The proposed RBPPNN information processing mechanism was extended to the time domain, and through learning time-varying signal training samples, achieved extraction, expression, and information association of time-varying signal characteristics, as well as direct classification. It can improve the deficiencies of existing neural networks, such as a complete large-scale training dataset is needed, and the information processing flow is complex. In this paper, the properties of the RBPPNN are analyzed and a specific learning algorithm is presented which synthesizes dynamic time warping, dynamic C-means clustering, and the mean square error algorithm. A series of 12-lead electrocardiogram (ECG) signals were used for classification testing of heart disease diagnosis results. The ECG classification accuracy across ten disease types was 75.52% and sinus arrhythmia was identified with an accuracy of 86.75%, verifying the effectiveness of the model and algorithm.

GPUs Outperform Current HPC and Neuromorphic Solutions in Terms of Speed and Energy When Simulating a Highly-Connected Cortical Model.

While neuromorphic systems may be the ultimate platform for deploying spiking neural networks (SNNs), their distributed nature and optimization for specific types of models makes them unwieldy tools for developing them. Instead, SNN models tend to be developed and simulated on computers or clusters of computers with standard von Neumann CPU architectures. Over the last decade, as well as becoming a common fixture in many workstations, NVIDIA GPU accelerators have entered the High Performance Computing field and are now used in 50 % of the Top 10 super computing sites worldwide. In this paper we use our GeNN code generator to re-implement two neo-cortex-inspired, circuit-scale, point neuron network models on GPU hardware. We verify the correctness of our GPU simulations against prior results obtained with NEST running on traditional HPC hardware and compare the performance with respect to speed and energy consumption against published data from CPU-based HPC and neuromorphic hardware. A full-scale model of a cortical column can be simulated at speeds approaching 0.5× real-time using a single NVIDIA Tesla V100 accelerator—faster than is currently possible using a CPU based cluster or the SpiNNaker neuromorphic system. In addition, we find that, across a range of GPU systems, the energy to solution as well as the energy per synaptic event of the microcircuit simulation is as much as 14× lower than either on SpiNNaker or in CPU-based simulations. Besides performance in terms of speed and energy consumption of the simulation, efficient initialization of models is also a crucial concern, particularly in a research context where repeated runs and parameter-space exploration are required. Therefore, we also introduce in this paper some of the novel parallel initialization methods implemented in the latest version of GeNN and demonstrate how they can enable further speed and energy advantages.

Motor primitives in space and time via targeted gain modulation in cortical networks.

Motor cortex (M1) exhibits a rich repertoire of neuronal activities to support the generation of complex movements. Although recent neuronal-network models capture many qualitative aspects of M1 dynamics, they can generate only a few distinct movements. Additionally, it is unclear how M1 efficiently controls movements over a wide range of shapes and speeds. We demonstrate that modulation of neuronal input–output gains in recurrent neuronal-network models with fixed architecture can dramatically reorganize neuronal activity and thus downstream muscle outputs. Consistent with the observation of diffuse neuromodulatory projections to M1, a relatively small number of modulatory control units provide sufficient flexibility to adjust high-dimensional network activity using a simple reward-based learning rule. Furthermore, it is possible to assemble novel movements from previously learned primitives, and one can separately change movement speed while preserving movement shape. Our results provide a new perspective on the role of modulatory systems in controlling recurrent cortical activity.

Constraints of Metabolic Energy on the Number of Synaptic Connections of Neurons and the Density of Neuronal Networks.

Neuronal networks in the brain are the structural basis of human cognitive function, and the plasticity of neuronal networks is thought to be the principal neural mechanism underlying learning and memory. Dominated by the Hebbian theory, researchers have devoted extensive effort to studying the changes in synaptic connections between neurons. However, understanding the network topology of all synaptic connections has been neglected over the past decades. Furthermore, increasing studies indicate that synaptic activities are tightly coupled with metabolic energy, and metabolic energy is a unifying principle governing neuronal activities. Therefore, the network topology of all synaptic connections may also be governed by metabolic energy. Here, by implementing a computational model, we investigate the general synaptic organization rules for neurons and neuronal networks from the perspective of energy metabolism. We find that to maintain the energy balance of individual neurons in the proposed model, the number of synaptic connections is inversely proportional to the average of the synaptic weights. This strategy may be adopted by neurons to ensure that the ability of neurons to transmit signals matches their own energy metabolism. In addition, we find that the density of neuronal networks is also an important factor in the energy balance of neuronal networks. An abnormal increase or decrease in the network density could lead to failure of energy metabolism in the neuronal network. These rules may change our view of neuronal networks in the brain and have guiding significance for the design of neuronal network models.

An integrated model for vent area design of hydrocarbon-air mixture explosion inside cubic enclosures with obstacles

This study aims to develop an integrated model - NFPA-68-BRANN model, which can be used to calculate the vent areas of cubic enclosures with obstacles. Seven experiments regarding vented explosion inside the obstructed enclosure are reviewed and applied to check the accuracy of two existing standards, i.e. the NFPA-68 2018 and the BS EN 14994:2007. Accordingly, the parameters to describe the flame development in the NFPA-68 2018 are amended by adopting the Bauwens model. Bayesian Regularization Artificial Neuron Network (BRANN) model presenting the non-linear relationship between the turbulent flame enhancement factor X and its affecting factors is subsequently developed. Eventually, the NFPA-68-BRANN model is generated by incorporating the BRANN model into the modified NFAP-68 2018. The accuracy of the NFPA-68-BRANN model is validated by using a series of the New Baker Test data.

VIOLA-A Multi-Purpose and Web-Based Visualization Tool for Neuronal-Network Simulation Output.

Neuronal network models and corresponding computer simulations are invaluable tools to aid the interpretation of the relationship between neuron properties, connectivity, and measured activity in cortical tissue. Spatiotemporal patterns of activity propagating across the cortical surface as observed experimentally can for example be described by neuronal network models with layered geometry and distance-dependent connectivity. In order to cover the surface area captured by today's experimental techniques and to achieve sufficient self-consistency, such models contain millions of nerve cells. The interpretation of the resulting stream of multi-modal and multi-dimensional simulation data calls for integrating interactive visualization steps into existing simulation-analysis workflows. Here, we present a set of interactive visualization concepts called views for the visual analysis of activity data in topological network models, and a corresponding reference implementation VIOLA (VIsualization Of Layer Activity). The software is a lightweight, open-source, web-based, and platform-independent application combining and adapting modern interactive visualization paradigms, such as coordinated multiple views, for massively parallel neurophysiological data. For a use-case demonstration we consider spiking activity data of a two-population, layered point-neuron network model incorporating distance-dependent connectivity subject to a spatially confined excitation originating from an external population. With the multiple coordinated views, an explorative and qualitative assessment of the spatiotemporal features of neuronal activity can be performed upfront of a detailed quantitative data analysis of specific aspects of the data. Interactive multi-view analysis therefore assists existing data analysis workflows. Furthermore, ongoing efforts including the European Human Brain Project aim at providing online user portals for integrated model development, simulation, analysis, and provenance tracking, wherein interactive visual analysis tools are one component. Browser-compatible, web-technology based solutions are therefore required. Within this scope, with VIOLA we provide a first prototype.

Delayed feedback control of phase synchronisation in a neuronal network model

The human cerebral cortex can be separated into cortical areas forming a clustered network structure. We build two different clustered networks, where one network is based on a healthy brain and the other according to a brain affected by a neurodegenerative process. Each cortical area has a subnetwork with small-world properties. We verify that both networks exhibit rich-club organisation and phase synchronisation. Due to the fact that neuronal synchronisation can be related to brain diseases, we consider the delayed feedback control as a method to suppress synchronous behaviours. In this work, it is presented that depending on the feedback parameters, intensity and time delay, phase synchronisation in both networks can be suppressed. Therefore, one of our main results is to show that delayed feedback control can be used to suppress undesired synchronous behaviours not only in the healthy brain, but also in the brain marked by neurodegenerative processes.

The impact of spike-frequency adaptation on balanced network dynamics.

A dynamic balance between strong excitatory and inhibitory neuronal inputs is hypothesized to play a pivotal role in information processing in the brain. While there is evidence of the existence of a balanced operating regime in several cortical areas and idealized neuronal network models, it is important for the theory of balanced networks to be reconciled with more physiological neuronal modeling assumptions. In this work, we examine the impact of spike-frequency adaptation, observed widely across neurons in the brain, on balanced dynamics. We incorporate adaptation into binary and integrate-and-fire neuronal network models, analyzing the theoretical effect of adaptation in the large network limit and performing an extensive numerical investigation of the model adaptation parameter space. Our analysis demonstrates that balance is well preserved for moderate adaptation strength even if the entire network exhibits adaptation. In the common physiological case in which only excitatory neurons undergo adaptation, we show that the balanced operating regime in fact widens relative to the non-adaptive case. We hypothesize that spike-frequency adaptation may have been selected through evolution to robustly facilitate balanced dynamics across diverse cognitive operating states.

INCF Program on Multiscale Modeling Workshop: From Cellular/Network Models to Tissue Simulation

The INCF Program on Multiscale Modeling (MSM) aims to promote the development of software infrastructure to facilitate neural modeling across many different scales. A new focus of the Program could be to support the transition from single neuron (network) modeling to neural tissue simulation. The aim would be to develop model description languages and the software infrastructure required to leverage existing molecular, neuron, and network models to build neural tissue models that accommodate both virtual tissue and reconstructed tissue modeling within a single framework. The aim of this workshop was to assess the current state of the field and available software infrastructure for tissue modeling from within the field of neuroscience and other fields in order to broadly define an outline of the development needed. The INCF will take the recommendations of this workshop to determine the need for subsequent actions and follow-up workshops, in order to direct future activities within the MSM Program, including the formation of new Program task force(s) as needed. In addition, The INCF aims to integrate these efforts with other INCF Programs such as those currently in progress in the Digital Brain Atlasing Program, Standards for Data Sharing Program and the Program on Ontologies of Neural Structures. The participants of the workshop suggested that the INCF could establish a working group to investigate how to describe spatial structures at the micro-scale, including the investigation of annotated mesh formats, mesh outputs from EM reconstruction, coordinate transforms, and markup schemes (Digital Atlasing, SBML). Participants noted that a thematic workshop on topic of ‘How to model when you do not know the parameters’ would be useful, with computational biologists included as participants. The participants also recommended that the INCF should announce an open call in the form of a competition for interested parties who want to commit to commenting on the state of the field for some of the use cases recommended during the discussion. The candidates should respond with a plan which will be reviewed against a set of criteria set forth during the workshop. The INCF should then hold followup workshops around the use cases with the output of these workshops being a set of recommendations for actions by the INCF. One identified use case linking different levels involves advancing the understanding of the relationship of sub cellular processes and cellular level electrical behavior. Since the workshop, a proposal has been developed further by the community for a workshop on the topic ‘Methods in bridging subcellular and cellular level models’.

Evidence for a Neuronal Network Model of the Cochlear Nucleus and Its Implication for Coding Strategies in Auditory Prostheses

Evidence for a Neuronal Network Model of the Cochlear Nucleus and Its Implication for Coding Strategies in Auditory Prostheses

Beta-Rhythm Oscillations and Synchronization Transition in Network Models of Izhikevich Neurons: Effect of Topology and Synaptic Type

Despite their significant functional roles, beta-band oscillations are least understood. Synchronization in neuronal networks have attracted much attention in recent years with the main focus on transition type. Whether one obtains explosive transition or a continuous transition is an important feature of the neuronal network which can depend on network structure as well as synaptic types. In this study we consider the effect of synaptic interaction (electrical and chemical) as well as structural connectivity on synchronization transition in network models of Izhikevich neurons which spike regularly with beta rhythms. We find a wide range of behavior including continuous transition, explosive transition, as well as lack of global order. The stronger electrical synapses are more conducive to synchronization and can even lead to explosive synchronization. The key network element which determines the order of transition is found to be the clustering coefficient and not the small world effect, or the existence of hubs in a network. These results are in contrast to previous results which use phase oscillator models such as the Kuramoto model. Furthermore, we show that the patterns of synchronization changes when one goes to the gamma band. We attribute such a change to the change in the refractory period of Izhikevich neurons which changes significantly with frequency.

Weak and Strong Connectivity Regimes for a General Time Elapsed Neuron Network Model

For large fully connected neuron networks, we study the dynamics of homogenous assemblies of interacting neurons described by time elapsed models. Under general assumptions on the firing rate which include the ones made in previous works [7, 8, 6], we establish accurate estimate on the long time behavior of the solutions in the weak and the strong connectivity regime both in the case with and without delay. Our results improve [7, 8] where a less accurate estimate was established and [6] where only smooth firing rates were considered. Our approach combines several arguments introduced in the above previous works as well as a slightly refined version of the Weyl's and spectral mapping theorems presented in [13, 4].

PENGGUNAAN MODEL JARINGAN SARAF TIRUAN (ARTIFICIAL NEURON NETWORK) UNTUK MEMPREDIKSI HASIL TANDAN BUAH SEGAR (TBS) KELAPA SAWIT BERDASAR CURAH HUJAN DAN HASIL TBS SEBELUMNYA

Abstrak. Untuk memprediksi hasil kelapa sawit 2018 di 4 kebun percobaan Pusat Penelitian Kelapa Sawit (Padang Mandarsah, Dalu-dalu, Bukit Sentang, dan Aek Pancur), maka disusun model jaringan saraf tiruan (JST). Data yang digunakan untuk menyusun model ini adalah hasil dan curah hujan bulanan pada 2013-2017. Keluaran model diperolah dari hubungan non-linear Autoregressive dengan masukkan eksternal (External/Exogenous input, NARX) yaitu curah hujan bulanan. Proses pemodelan meliputi training model menggunakan data 2013-2015, validasi model menggunakan data 2016, dan pengujian model menggunakan data 2017. Dari hasil pengujian model diperolah arsitektur model JST n-d-h-o (peubah input,n ; d-tapped delayed , d, node tersembunyi, h; lapisan keluaran, o) yang sesuai untuk kebun Padang Mandarsah , Dalu-dalu, Bukit Sentang, dan Aek Pancur berturut-turut adalah 2-3-4-1, 2-24-5-1, 2-24-10-1, dan 2-3-5-1 dengan hubungan keeratan (r) antara keluaran model dengan data aktual berturut-turut 0,84; 0,74; 0,84; dan 0,86.

An Oscillatory Neural Autoencoder Based on Frequency Modulation and Multiplexing.

Oscillatory phenomena are ubiquitous in the brain. Although there are oscillator-based models of brain dynamics, their universal computational properties have not been explored much unlike in the case of rate-coded and spiking neuron network models. Use of oscillator-based models is often limited to special phenomena like locomotor rhythms and oscillatory attractor-based memories. If neuronal ensembles are taken to be the basic functional units of brain dynamics, it is desirable to develop oscillator-based models that can explain a wide variety of neural phenomena. Autoencoders are a special type of feed forward networks that have been used for construction of large-scale deep networks. Although autoencoders based on rate-coded and spiking neuron networks have been proposed, there are no autoencoders based on oscillators. We propose here an oscillatory neural network model that performs the function of an autoencoder. The model is a hybrid of rate-coded neurons and neural oscillators. Input signals modulate the frequency of the neural encoder oscillators. These signals are then multiplexed using a network of rate-code neurons that has afferent Hebbian and lateral anti-Hebbian connectivity, termed as Lateral Anti Hebbian Network (LAHN). Finally the LAHN output is de-multiplexed using an output neural layer which is a combination of adaptive Hopf and Kuramoto oscillators for the signal reconstruction. The Kuramoto-Hopf combination performing demodulation is a novel way of describing a neural phase-locked loop. The proposed model is tested using both synthetic signals and real world EEG signals. The proposed model arises out of the general motivation to construct biologically inspired, oscillatory versions of some of the standard neural network models, and presents itself as an autoencoder network based on oscillatory neurons applicable to time series signals. As a demonstration, the model is applied to compression of EEG signals.

Criteria on Balance, Stability, and Excitability in Cortical Networks for Constraining Computational Models.

During ongoing and Up state activity, cortical circuits manifest a set of dynamical features that are conserved across these states. The present work systematizes these phenomena by three notions: excitability, the ability to sustain activity without external input; balance, precise coordination of excitatory and inhibitory neuronal inputs; and stability, maintenance of activity at a steady level. Slice preparations exhibiting Up states demonstrate that balanced activity can be maintained by small local circuits. While computational models of cortical circuits have included different combinations of excitability, balance, and stability, they have done so without a systematic quantitative comparison with experimental data. Our study provides quantitative criteria for this purpose, by analyzing in-vitro and in-vivo neuronal activity and characterizing the dynamics on the neuronal and population levels. The criteria are defined with a tolerance that allows for differences between experiments, yet are sufficient to capture commonalities between persistently depolarized cortical network states and to help validate computational models of cortex. As test cases for the derived set of criteria, we analyze three widely used models of cortical circuits and find that each model possesses some of the experimentally observed features, but none satisfies all criteria simultaneously, showing that the criteria are able to identify weak spots in computational models. The criteria described here form a starting point for the systematic validation of cortical neuronal network models, which will help improve the reliability of future models, and render them better building blocks for larger models of the brain.

A realistic neuronal network and neurovascular coupling model for the study of multivariate directed connectivity in fMRI data.

The use of Multivariate Granger Causality (MVGC) in estimating directed Blood-Oxygen-Level- Dependant (BOLD) connectivity is still controversial. This is mostly due to the short data Ienghts typically available in func- tional MRI (fMRI) acquisitions, to the very nature of the BOLD acquisition strategy (which yields extremely low signal- to-noise-ratio) and importantly to the fact that neuronal activi- ty is convolved with a slow-varying haemodynamic response function (HRF) which therefore generates a temporal confound which is arduous to account for when basing MVGC estimates on vector autoregressive models (VAR). In this paper, we em- ploy realistic complex network models based on Izhikevich neuronal populations, interlinked by realistic neuronal fiber bundles which exert compounded directed influences and cas- cade into Baloon-model-like neurovascular coupling, to explore and validate the MVGC approach to directed connectivity es- timation in realistic fMRI conditions and in a complex directed network setting. In particular, we show in silico that the top 1 percentile of a BOLD connectivity matrix estimated with MVGC from BOLD data similar to the one provided by the Human Connectome Project (HCP) has a Positive Predictive Value very close to 1, hence corroborating the evidence that the "strongest" connections can be safely studied with this method in fMRI.