In this paper a chaotic spiking neuron is presented and its response characteristics to various periodic inputs are analyzed. A return map which can analytically describe the dynamics of the neuron is derived. Using the map, it is theoretically shown that a set of neurons can encode various periodic inputs into a set of spike-trains in such a way that a spike density of a summation of the spike-trains can approximate the waveform of the input. Based on the theoretical results, some potential applications of the presented neuron are also discussed. Using a prototype circuit, typical encoding functions of the neuron are confirmed by experimental measurements.

In this paper, we propose an analog CMOS circuit which achieves spiking neural networks with spike-timing dependent synaptic plasticity (STDP). In particular, we propose a STDP circuit with symmetric function for the first time, and also we demonstrate associative memory operation in a Hopfield-type feedback network with STDP learning. In our spiking neuron model, analog information expressing processing results is given by the relative timing of spike firing events. It is well known that a biological neuron changes its synaptic weights by STDP, which provides learning rules depending on relative timing between asynchronous spikes. Therefore, STDP can be used for spiking neural systems with learning function. The measurement results of fabricated chips using TSMC 0.25 µm CMOS process technology demonstrate that our spiking neuron circuit can construct feedback networks and update synaptic weights based on relative timing between asynchronous spikes by a symmetric or an asymmetric STDP circuits.

This paper studies a simple spiking oscillator having piecewise constant vector field. Repeating vibrate-and-fire dynamics, the system exhibits various spike-trains and we pay special attention to chaotic spike-trains having line-like spectrum in distribution of inter-spike intervals. In the parameter space, existence regions of such phenomena can construct infinite window-like structures. The system has piecewise linear trajectory and we can give theoretical evidence for the phenomena. Presenting a simple test circuit, typical phenomena are confirmed experimentally.

This letter proposes a neurobiological approach for action recognition. In this approach, actions are represented by a visual-neuron feature (VNF) based on a quantitative model of object representation in the primate visual cortex. A supervised classification technique is then used to classify the actions. The proposed VNF is invariant to affine translation and scaling of moving objects while maintaining action specificity. Moreover, it is robust to the deformation of actors. Experiments on publicly available action datasets demonstrate the proposed approach outperforms conventional action recognition models based on computer-vision features.

This letter studies response of a chaotic spiking oscillator to chaotic spike-train inputs. The circuit can exhibits a variety of synchronous/asynchronous phenomena and we show an interesting phenomenon "consistency": the circuit can exhibit random response that is identical in steady steady state for various initial values. Presenting a simple test circuit, the consistency is confirmed experimentally.

This paper studies encoding/decoding function of artificial spiking neurons. First, we investigate basic characteristics of spike-trains of the neurons and fix parameter value that can minimize variation of spike-train length for initial value. Second we consider analog-to-digital encoding based upon spike-interval modulation that is suitable for simple and stable signal detection. Third we present a digital-to-analog decoder in which digital input is applied to switch the base signal of the spiking neuron. The system dynamics can be simplified into simple switched dynamical systems and precise analysis is possible. A simple circuit model is also presented.

Further development of a network based on the Inverse Function Delayed (ID) model which can recall temporal sequences of patterns, is proposed. Additional advantage is taken of the negative resistance region of the ID model and its hysteretic properties by widening the negative resistance region and letting the output of the ID neuron be almost instant. Calling this neuron limit ID neuron, a model with limit ID neurons connected pairwise with conventional neurons enlarges the storage capacity and increases it even further by using a weightmatrix that is calculated to guarantee the storage after transforming the sequence of patterns into a linear separation problem. The network's tolerance, or the model's ability to recall a sequence, starting in a pattern with initial distortion is also investigated and by choosing a suitable value for the output delay of the conventional neuron, the distortion is gradually reduced and finally vanishes.

Two simple but useful methods, called the deep learning methods, for making multilayer neural networks tolerant to multiple link-weight and neuron-output faults, are proposed. The methods make the output errors in learning phase smaller than those in practical use. The abilities of fault-tolerance of the multilayer neural networks in practical use, are analyzed in the relationship between the output errors in learning phase and in practical use. The analytical result shows that the multilayer neural networks have complete (100%) fault-tolerance to multiple weight-and-neuron faults in practical use. The simulation results concerning the rate of successful learnings, the ability of fault-tolerance, and the learning time, are also shown.

This paper studies a spiking neuron circuit with triangular base signal. The circuit can output rich spike-trains and the dynamics can be analyzed using a one-dimensional piecewise linear map. This system exhibits period doubling bifurcation, tangent bifurcation, super-stable periodic orbit bifurcation and so on. These phenomena can be characterized based on the inter-spike intervals. Using the maps, we can analyze the phenomena precisely. By presenting a simple test circuit, typical phenomena are confirmed experimentally.

The digital spiking neuron (DSN) consists of digital state cells and behaves like a simplified neuron model. By adjusting wirings among the cells, the DSN can generate spike-trains with various characteristics. In this paper we present a theorem that clarifies basic relations between change of wirings and change of characteristics of the spike-train. Also, in order to explore learning potential of the DSN, we propose a learning algorithm for generating spike-trains that are suited to an application example. We then show significances and basic roles of the presented theorem in the learning dynamics.

An integrate-and-fire-type spiking feedback network is discussed in this paper. In our spiking neuron model, analog information expressing processing results is given by the relative relation of spike firing. Therefore, for spiking feedback networks, all neurons should fire (pseudo-)periodically. However, an integrate-and-fire-type neuron generates no spike unless its internal potential exceeds the threshold. To solve this problem, we propose negative thresholding operation. In this paper, this operation is achieved by a global excitatory unit. This unit operates immediately after receiving the first spike input. We have designed a CMOS spiking feedback network VLSI circuit with the global excitatory unit for Hopfield-type associative memory. The circuit simulation results show that the network achieves correct association operation.

Cell assembly is one of explanations of information processing in the brain, in which an information is represented by a firing space pattern of a group of plural neurons. On the other hand, effectiveness of neural network has been confirmed in pattern recognition, system control, signal processing, and so on, since the back propagation learning was proposed. In this study, we propose a new network structure with affordable neurons in the hidden layer of the feedforward neural network. Computer simulated results show that the proposed network exhibits a good performance for the back propagation learning. Furthermore, we confirm the proposed network has a good generalization ability.

In this paper, we propose a modified bursting neuron model, which is a natural extension of an original one proposed by the author et al. We will show that chaotic bursts appear in the modified model though there exhibit quasi-periodic bursts in the original one. Moreover, we will show that such chaotic bursts appear by breaking down a pair of invariant closed curves, which is generated by a Hopf bifurcation for a pair of two-periodic points.

This letter studies a simple nonautonomous chaotic circuit constructed by adding an impulsive switch to the RCL circuit. The switch operation depends on time and on state variable through a refractory threshold. The circuit exhibits various chaotic attractors, periodic attractors and related bifurcation phenomena. The dynamics can be analyzed using 1-D return map focusing on the time-dependent switching moments. Using a simple test circuit model typical phenomena are verified in PSPICE simulations.

This paper presents pulse-coupled two bifurcating neurons. The single neuron is represented by a spike position map and the coupled neurons can be represented by a composition of the spike position maps. Using the composite map, we can analyze basic bifurcation phenomena and can find some interesting phenomena that are caused by the pulse-coupling and are impossible in the single neuron. Presenting a simple test circuit, typical phenomena are confirmed experimentally.

We present mutually pulse-coupled two relaxation oscillators having refractoriness. The system can be implemented by a simple electrical circuit, and various periodic synchronization phenomena can be observed experimentally. The phenomena are characterized by a ratio of phase locking. Using a return map having a trapping window, the ratio can be analyzed in a parameter subspace rigorously. We then clarify effects of the refractoriness on the pulse coding ability of the system.

The hardware implementation of a neural network model using stochastic logic has been able to integrate numerous neuron units on a chip. However, the limitation of applications occurred since the stochastic neurosystem could execute only discrete-time dynamics. We have contrived a neuron model with continuous-time dynamics by using stochastic calculations. In this paper, we propose the circuit design of a new neuron circuit, and show the fabricated neurochip comprising 64 neurons with experimental results. Furthermore, a new asynchronous updating method and a new activation function circuit are proposed. These improvements enhance the performance of the neurochip greatly.

We present master-slave pulse-coupled bifurcating neurons having refractoriness. The system can exhibit various phenomena, e. g. , periodic and chaotic in-phase synchronizations, and periodic out-of-phase synchronization. We clarify local stabilities of the phenomena and a sufficient condition for the in-phase synchronization. It is suggested that bifurcations of the synchronization phenomena may relate to detection of a master parameter, and the refractoriness may relate to control of the detection accuracy. Using a simple test circuit, typical phenomena are verified in the laboratory.

Utilizing a macromodel which calculates the floating gate potential by combining resistances and dependent voltage and current sources, DC transfer characteristics for multi-input neuron MOS inverters and for those in the neuron MOS full adder circuit are simulated both at room temperature and at 77 K. Based on the simulated results, low temperature circuit failures are discussed. Furthermore, circuit design parameter optimization both for low and room temperature operations is described.

In this paper, we propose the CMOS implementation of neuron models for an artificial auditory neural network. We show that when voltage is added directly to the control terminal of the basic circuit of the hardware neuron model, a change in the output firing is observed. Next, based on this circuit, a circuit that changes with time is added to the control terminal of the basic circuit of the hardware neuron model. As a result, a neuron model is constructed with ON firing, adaptation firing, and repetitive firing using CMOS. Furthermore, an improved circuit of a neuron model with OFF firing using CMOS which has been improved from the previous model is also constructed.