The biologically discovered intrinsic plasticity (IP) learning rule, which changes the excitability of an individual neuron by adaptively turning firing threshold, has been shown to be crucial for efficient information processing. However, this rule needs extra time updating operations at each step, causing energy consumption and reducing computational efficiency. event-driven or spike-based coding strategy spiking neural networks (SNNs), i.e., neurons will only active if driven continuous...

Topology identification of a network has received great interest for the reason that study on many key properties assumes special known topology. Different from recent similar works in which evolution all nodes complex need to be received, this brief presents novel criterion identify topology coupled FitzHugh-Nagumo (FHN) neurobiological by receiving membrane potentials only fraction neurons. Meanwhile, although incomplete information is neurons including and recovery variables are traced....

The response of three coupled FitzHugh-Nagumo neurons, under Gaussian white noise, to a subthreshold periodic signal is studied in this paper. By combining the canard dynamics, chemical coupling, and stochastic resonance together, information transfer neural system investigated. We find that synaptic coupling more efficient than well-known linear (gap junction) for local input, i.e., only one neurons subject signal. This weak input common biological systems sake low energy consumption.

Most network models for neural behavior assume a predefined topology and consist of almost identical elements exhibiting little heterogeneity. In this paper, we propose self-organized consisting heterogeneous neurons with different behaviors or degrees excitability. The synaptic connections evolve according to the spike-timing dependent plasticity mechanism finally sparse active-neuron-dominant structure is observed. That is, strong are mainly distributed synapses from active inactive ones....

In recent years, the stock market plays an important role, which has attracted more and attentions. The key problem of prediction is how to design a method improve performance. As we know, biggest challenge that time series essentially dynamic, nonlinear, complicated, nonparametric chaotic. this paper, propose novel predict closing price based on deep belief networks (DBNs) with intrinsic plasticity. experiments, in S&P 500 used examine back propagation algorithm for output training make...

The distal apical dendrites of layer 5 pyramidal neurons receive cortico-cortical and thalamocortical top-down feedback inputs, as well local recurrent inputs. A prominent source inhibition in the neocortical circuit is somatostatin-positive Martinotti cells, which preferentially target cells. These electrically coupled cells can fire synchronously at various frequencies, including over a relatively slow range (5∼30 Hz), thereby imposing oscillatory on tuft dendrites. We examined how such...

Long-term synaptic plasticity induced by neural activity is of great importance in informing the formation connectivity and development nervous system. It reasonable to consider self-organized networks instead prior imposition a specific topology. In this paper, we propose novel network evolved from two stages learning process, which are respectively guided experimentally observed rules, i.e. spike-timing-dependent (STDP) mechanism burst-timing-dependent (BTDP) mechanism. Due existence...

Neuronal avalanche is a spontaneous neuronal activity which obeys power-law distribution of population event sizes with an exponent –3/2. It has been observed in the superficial layers cortex both invivo and invitro. In this paper, we analyze information transmission novel self-organized neural network active-neuron-dominant structure. avalanches can be appropriate input intensity. We find that process learning via spike-timing dependent plasticity dramatically increases complexity...

Recently, echo state network (ESN) has attracted a great deal of attention due to its high accuracy and efficient learning performance. Compared with the traditional random structure classical sigmoid units, simple circle topology leaky integrator neurons have more advantages on reservoir computing ESN. In this paper, we propose new model ESN both units. By comparing prediction capability Mackey-Glass chaotic time series four models: ESN, find that our shows significantly better performance...

Homomorphic encryption technology can settle a dispute of data privacy security in cloud environment, but there are many problems the process access which is encrypted by homomorphic algorithm cloud. In this paper, on premise attribute encryption, we propose fully encrypt scheme based with LSSS matrix. This supports fine-grained cum flexible control along "Query-Response" mechanism to enable users efficiently retrieve desired from servers. addition, should support considerable flexibility...

In recent years, an increasing number of studies have demonstrated that networks in the brain can self-organize into a critical state where dynamics exhibit mixture ordered and disordered patterns. This branching phenomenon is termed neuronal avalanches. It has been hypothesized homeostatic level balanced between stability plasticity this may be optimal for performing diverse neural computational tasks. However, region high performance narrow sensitive spiking (SNNs). paper, we investigated...

Echo state networks (ESNs) with multi-clustered reservoir topology perform better in computing and robustness than those random topology. However, these ESNs have a complex topology, which leads to difficulties generation. This study focuses on the generation problem when ESN is used environments sufficient priori data available. Accordingly, data-driven multi-cluster algorithm proposed. The proposed are evaluate reservoirs by calculating precision standard deviation of ESNs. produced using...

Liquid State Machine (LSM) is a biologically plausible neural network model for real-time computing on time-varying inputs, which shown to be beneficial perform computational tasks like pattern classification. The LSM uses spiking neurons with dynamic synapses projects inputs into high-dimensional space, facilitating subsequent linear recognition In this paper, we present two different methods improve in classification from perspectives of spatial integration and temporal integration. We...

Intrinsic plasticity (IP) is an unsupervised, self-adaptive, local learning rule that was first found in biological nerve cells, and has been shown to be able maximize neuronal information transmission entropy. In this article, we propose a soft-reset leaky integrate-and-fire (LIF) model, spiking neuron model based on widely used LIF neurons, with new IP optimizes the membrane potential state exponentially distributed. Previous studies have generally such as expected firing rate target...

In this article, we present an echo state network (ESN)-based tracking control approach for a class of affine nonlinear systems. Different from the most existing neural-network (NN)-based methods that are focused on feedforward NN, proposed method adopts bioinspired recurrent NN fusing with multiple cluster and intrinsic plasticity (IP) to deal modeling uncertainties coupling nonlinearities in The key features work can be summarized as follows: 1) is built upon ESN embedded multiclustered...

For ensuring the safe operation of IGBT, Artificial Intelligence technology can be used to predict life IGBT. Aiming at problem that remaining prediction method IGBT lacks accuracy due difficulty in selecting parameters LSTM model, a WOA is proposed optimize model. algorithm number hidden neurons and learning rate LSTM, which provides better for training. The model WOA-LSTM residual established by Python, compared with LSTM. experimental results show RMSE 0.1838, MAE 0.1483, MAPE 0.0103, has...