37 results about "Artificial neuron" patented technology

The application provides a neuron-like transistor device, which comprises a substrate, a buried oxygen layer and a channel layer which are sequentially stacked on the substrate, a gate oxide layer and a gate electrode on the channel layer, and a source electrode and a drain electrode which are respectively located on both sides of the channel layer. The gate electrode is used for applying a front gate voltage as an input signal, the drain electrode is used for applying a constant input current, the voltage of the drain electrode is used as an output signal, and the input signal is changed in a step-by-step manner from small to large. The neuron nonlinear activation of a single device is realized, and it is possible to build a complete and programmable artificial neuron by using a single or few transistors. The neuron-like transistor device has a simple structure, and a manufacturing process which is suitable for a CMOS manufacturing process. The measurement method of the neuron-like transistor device is simple, and an electrode does not need to be additionally introduced, so that the simulation of the neuron characteristics can be realized without additional device design, the integration density of the transistors in a chip is improved, and the neuron-like transistor device is suitable for the integration of a large-scale circuit.

An artificial neuron includes a first capacitive node of application of a membrane potential of the neuron. A first transistor is configured to discharge the first capacitive node. A second capacitive node is driven according to the membrane potential and delivers a potential for controlling the first transistor. A second transistor is configured to discharge the second capacitive node. The second transistor is controlled according to a potential present at the second capacitive node.

The application belongs to the field of artificial neuron devices, in order to solve the problems of single function, complex manufacturing process and inability to adapt to mass production of the existing artificial neuron devices, the application provides a preparation method of a micro multifunctional artificial neuron device based on a two-dimensional WSe2 transistor, comprising the following steps: preparing a single-layer / few-layer selenium disulfide thin channel; evaporating Se / Au electrodes to manufacture a P-type WSe2 transistor; evaporating Bi / Au electrodes to manufacture an N-type WSe2 transistor and connecting by adopting a CMOS structure; the application also tests the electrical performance of the prepared device and the input and output curves of the activation function thereof, simulates various indexes and obtains the accuracy of the simulation results, and proves that the application can be applied to realize the activation function and the artificial synapse function.

The present application relates to the field of artificial neuromorphics, and provides an artificial neuron circuit based on a volatile threshold switching device. A membrane potential generation circuit is used for generating different membrane potentials under different input currents and different external reset voltages; a slow variable generation circuit is used for generating a slow variable; a membrane potential reset circuit comprises a volatile threshold switching device and an external reset voltage excitation source; the external reset voltage excitation source is used for providing different external reset voltages, so that the membrane potential and the slow variable are changed, and a spiking behavior of biological neurons is simulated; and the volatile threshold switching device is used for realizing a membrane potential reset function. Compared with traditional CMOS-circuit-based neurons, the neuron circuit provided by the present application requires a greatly reduced number of transistors and reduced hardware and power consumption overhead, has the characteristics of simple structure, high flexibility and rich functions, and is beneficial to large-scale integration in a hardware pulse neural network.

The embodiment of the application discloses a kind of bridge artificial neuron unit composition's bionic hearing positioning system based on memristor circuit, it is related to neural morphic computing and artificial intelligence hardware field.In the present application: bridge artificial neuron unit includes the first memristor analog circuit connected with first capacitor and first load resistance, and the second memristor analog circuit connected with second capacitor and second load resistance, which are connected by first bridge capacitor, and coupled with external circuit by second bridge capacitor.The unit can simulate the pulse emission and refractory behavior of biological neuron.A plurality of the unit can be connected in series to build a one-way pulse propagation chain, forming a pulse propagation network.Based on the network, a coincidence detector can detect the small time difference of multiple input signals and be applied to a bionic hearing system to achieve sound source positioning.The present application has simple structure, high degree of modularity, and can realize high-reliability analog pulse propagation and high-precision coincidence detection.

The present invention provides a method for calculating decision variables. A dummy layer is added at an input layer of a trained neural network predictive model. The dummy layer includes a plurality of artificial neurons respectively connected to a corresponding input terminal of the input layer for the trained predictive model by a newly established link. The input value of each artificial neuron is set to 1, the bias value of the activation function is set to 0, and the output of the activation function is set to 1 when the input of the activation function is 1. The initial weight value of the newly established link is selected and set, and the weight values can be considered as decision variables, wherein the weight values can have ranges or other inter-conditional restrictions. The optimal solution is obtained using the optimizer built in a general machine-learning platform when the parameters of the trained predictive model are frozen, and only the weight values of the newly established links are adjusted. The training objective is set so that the output of the parameter predictive model matches the desired target result. At the end of the training, the weight values of the newly established link new are the feasible input decision variables. This invention allows the effective use of the general machine learning platforms and the built-in methods to find the optimal input parameters that achieve the expected results.

The present invention is synthetic neurons and networks with feedback quantum tunneling memristors designed to mimic biological neurons. The memristor hardware is formed from a 4.2 nm thick layer of atomic layer deposited ionic hafnium oxide and niobium metal and inserted in the positive and negative feedback of an analog spiking oscillator. When operated at room / warm temperatures, these memories have memristive properties and enable the artificial neuron circuits to produce adaptive spiking behavior. When neural networks are formed of the synthetic neurons, pronounced hybrid chaotic / non-chaotic modes with increased complexity are observed and itinerant behavior emerges. Cryogenic cooling of the memristors into the superconducting Josephson tunneling regime at 8.1 K reveals the influence of quantum control effects.

The application is a kind of bionic nerve circuit model based on artificial synapse device. The circuit model is composed of n multi-stage connections of synapse-LIF artificial neuron devices. The composition of the synapse-LIF artificial neuron device includes synapse-like device and cytoplasm-like device. The synapse-like device includes artificial synapse device, voltage dividing resistor and voltage control switch. The cytoplasm-like device includes one comparator. Through different combinations of the four devices, namely excitatory synapse-like device, inhibitory synapse-like device, excitatory cytoplasm-like device and inhibitory cytoplasm-like device, the devices of six circuits, including feedforward excitation circuit, feedforward inhibition circuit, feedback inhibition circuit, lateral inhibition circuit, disinhibition circuit and mutual inhibition circuit, are formed. The application realizes the functions of perception afferent, information integration and motor control in the motor nervous system through the circuit model group, which is helpful for the repair or reconstruction of damaged motor conduction circuit.

A system including a supply valve disposed between a chamber and a pressure source, a discharge valve disposed between the chamber and an external atmosphere, a first control unit, and a second control unit. The first control unit coupled with the supply valve by a first switch and with the discharge valve by a second switch. The first control unit outputting signals to the first and second switches to control the supply and discharge valves. The second control unit coupled with the discharge valve by a third switch and a fourth switch, the second control unit outputting signals to the third and fourth switches to control the supply and discharge valves. The first control unit may include a first microcontroller to control the signals of the first control unit using an artificial intelligence (AI) neural network having artificial neurons arranged in layers and connected with each other by connections.

A hybrid neuromorphic computing device is provided, in which artificial neurons include light-emitting devices that provide weighted sums of inputs as light output. The output is detected by a photodetector and converted to an electrical output. Each neuron may receive output from one or more other neurons as initial input. Interconnects between neurons may be optical, electrical, or a combination thereof. The neurons also may provide imaging sensor and / or display capabilities.

A system and method for computing in memory with artificial neurons. According to an embodiment of the present disclosure, there is provided a system, including: a computer-readable memory; a neuron processing element communicatively connected to the computer-readable memory, the neuron processing element including: a plurality of configurable processing circuits each having a plurality of outputs and a plurality of inputs; and a network connecting one or more of the outputs of the configurable processing circuits to one or more of the inputs of the configurable processing circuits, each of the configurable processing circuits including: an artificial neuron having a plurality of inputs; and a register connected to the inputs of the artificial neuron.

A system for detecting the surroundings of a vehicle includes a sensor module and a plurality of computing devices which are spatially separated from one another and which are configured to process the detected surroundings data by way of an artificial neural network. The artificial neural network includes multiple layers of artificial neurons. Each computing device implements a partial number of the layers.

The application provides a wide-area refresh rate display driving system based on a neuromorphic device and a working method thereof, and belongs to the technical field of display, and comprises an input module, a control module and a light-emitting module; the control module comprises an enabling control module and a brightness control module which are connected with each other, the input module is connected with the enabling control module, and the light-emitting module is connected with the brightness control module; the input module splices and inputs an input enabling signal and an input data signal into the control module; the enabling control module adopts an artificial neuron device, and a single artificial neuron device is switched between on and off states according to the input enabling signal; the brightness control module adopts an artificial synapse device, and a single artificial synapse device is transformed among a plurality of different conductance states according to the input data signal; and the light-emitting module performs light-emitting display under the regulation and control of the brightness control module. The application can not only realize pixel-level control through an artificial neuron, but also realize wide-area refresh rate through an artificial synapse.

The invention relates to an event-driven spiking neural network system (100) and a method for detecting a physiological condition of a person based on a detected physiological signal of the person, the system comprising at least the following components:At least one sensor (111-1) configured and arranged to detect a physiological signal and to convert the physiological signal in a sensor signal (112) indicative of the physiological signal,A signal conversion module (120) configured and arranged to receive the sensor signal (112) from the at least one sensor (111-1) and to convert the sensor signal (112) in at least one time series of discrete events,An artificial neuron population (140) comprising a plurality of artificial event-driven spiking neurons (131-1, 131-N) arranged in an event-driven spiking neural network, wherein the neuron population (140) is configured and arranged to receive events, wherein the neuron population (140) is arranged to recognize the physiological condition of the person based on the received events, wherein the neuron population (140) is configured to provide one or more processed time series of events to,A condition detection module (150) arranged and configured to receive the events from the neuron population (140) and to output a trigger signal (180), when the events received from the artificial neuron population (140) indicate that the physiological signal comprises a feature indicative of the physiological condition.

Training large neural networks on big datasets requires significant computational resources and time. Transfer learning reduces training time by pre-training a base model on one dataset and transferring the knowledge to a new model for another dataset; while current choices of transfer learning algorithms are limited, biological neural networks (BNNs) are adept at rearranging themselves to tackle completely different problems using transfer learning. Taking advantage of BNNs, an artificial neural network (ANN) with dynamic transfer learning capability is transferable to any other network architecture and can accommodate many datasets. The ANN includes artificial neurons and artificial glial cells distributed within an N-dimensional space; connections are formed between pairs of artificial neurons that meet certain criteria. In an optogenetics implementation, machine learning models such as the disclosed ANN are implemented on real BNNs to decrease power consumption.

The application provides a neuron-like transistor device, which comprises a substrate, a buried oxygen layer and a channel layer which are sequentially stacked on the substrate, a gate oxide layer and a gate electrode on the channel layer, and a source electrode and a drain electrode which are respectively located on both sides of the channel layer. The gate electrode is used for applying a front gate voltage as an input signal, the drain electrode is used for applying a constant input current, the voltage of the drain electrode is used as an output signal, and the input signal is changed in a step-by-step manner from small to large. The neuron nonlinear activation of a single device is realized, and it is possible to build a complete and programmable artificial neuron by using a single or few transistors. The neuron-like transistor device has a simple structure, and a manufacturing process which is suitable for a CMOS manufacturing process. The measurement method of the neuron-like transistor device is simple, and an electrode does not need to be additionally introduced, so that the simulation of the neuron characteristics can be realized without additional device design, the integration density of the transistors in a chip is improved, and the neuron-like transistor device is suitable for the integration of a large-scale circuit.

An apparatus comprises an artificial neuron including a magnetic domain wall racetrack (racetrack), a first magnetic tunnel junction, a second magnetic tunnel junction, a first terminal, and a second terminal configured to cooperate to reliably create, propagate, readout, and reset a magnetic domain over multiple continuous operational cycles. The racetrack provides a propagation path for the magnetic domain. The first magnetic tunnel junction, coupled to the racetrack proximate to the first terminal, nucleates the magnetic domain in the racetrack. The first terminal is located proximate to one end of the racetrack, and receives a train of input voltage pulses that drive the magnetic domain along the racetrack. The second magnetic tunnel junction, coupled to the racetrack proximate to the second terminal, fires in response to passage of the magnetic domain. The second terminal, located proximate to another end of the racetrack, ejects the magnetic domain from the racetrack.

Artificial neurons are disclosed. An artificial neuron includes a first capacitance node to which a membrane potential of the neuron is applied. A first transistor is configured to discharge the first capacitance node. A second capacitance node is driven in accordance with the membrane potential and communicates a potential for controlling the first transistor. A second transistor is configured to discharge the second capacitance node. The second transistor is controlled in accordance with a potential present at the second capacitance node.

The invention discloses a Schmitt trigger, an artificial neuron circuit and a neuromorphic calculation device.The Schmitt trigger comprises a flexible substrate, a gate electrode, a biocompatible gate dielectric layer, a P-type organic semiconductor and an N-type organic semiconductor, complementary connection is achieved through three top electrodes, the capacitance value of the gate dielectric layer changes along with an electric field, and the gate dielectric layer has the hysteresis characteristic; the device has double threshold voltages in a single structure, and a Schmidt trigger function can be realized without a complex circuit. On the basis, the invention further provides an artificial neuron circuit, and the electric leakage integration-release behavior of the biological neurons is simulated by integrating the Schmitt trigger, the membrane capacitor and the phase inverter. The device has the advantages of simple structure, low power consumption, low preparation cost, excellent flexibility and biocompatibility and the like, and is suitable for an implantable bioelectronic and high-energy-efficiency neuromorphic computing system.

The invention provides an agricultural supply chain production place field market transaction generation digital-intelligent quotation technology and an AI method, and relates to the technical field of agricultural supply chain transaction data resource utilization research and development, AI decision making and transaction price artificial intelligence prediction. The method is a digital intelligent technology innovation aiming at an agricultural supply chain production place field end product marketing transaction activity. The method mainly comprises the following steps: a transaction vector metadata data processing [3: 3: 3: + / -1] mixed structure of an artificial intelligence model of a digital intelligence technology: vector metadata interconnection of an input layer, hidden layers I and II and an output layer; the AI system physical assembly comprises an information acquisition module, a data processing and representing module, an intelligent calculation module, a feedback and optimization module, a multi-stage transaction supporting module and a modular interface, and replacement and upgrading of the calculation module, the data processing module and the transaction supporting module are achieved. And the artificial neuron performs reverse propagation training and weight adjustment on the model by using a BP algorithm so as to perform continuous optimization.

Systems, methods, apparatuses, and computer program products for neural networks. In accordance with some example embodiments, an operational neuron model may comprise an artificial neuron comprising a composite nodal operator, a pool-operator, and an activation function operator. The nodal operator may comprise a linear function or non-linear function. In accordance with certain example embodiments, a generative neuron model may include a composite nodal-operator generated during the training using Taylor polynomial approximation without restrictions. In accordance with various example embodiments, a self-organized operational neural network (Self-ONN) may include one or more layers of generative neurons.

The invention discloses an LIF neuron circuit based on a self-powered threshold switch memristor and a preparation method of the LIF neuron circuit. The LIF neuron circuit comprises a friction nanometer generator, a rectifier bridge, a threshold TS memristor, a fixed value resistor and a capacitor. Dependence of a traditional artificial neuron on an external power supply and overall energy consumption of a system are remarkably reduced, meanwhile, key dynamic characteristics such as leakage, integration, distribution and delay of a biological neuron are efficiently simulated through cooperative charging and discharging behaviors of the memristor and the capacitor, a stimulation intensity-dependent frequency coding function is shown, and the simulation performance of the neuron is improved. And an effective hardware implementation scheme is provided for constructing a low-power-consumption and high-integration-level sensing and computing integrated neuromorphic system.

The application discloses a kind of cold-rolled thin strip steel plastic strain ratio on-line detection method and system, the cold-rolled thin strip steel plastic strain ratio on-line detection method includes the following steps: establishing the artificial neuron network algorithm model of cold-rolled thin strip steel;The plastic strain ratio of cold-rolled thin strip steel is obtained by artificial method;The artificial neuron network algorithm model of cold-rolled thin strip steel is modelled training;The artificial neuron network algorithm model of cold-rolled thin strip steel is used for on-line detection.The cold-rolled thin strip steel plastic strain ratio on-line detection method is applied to the comprehensive electromagnetic detection to running strip steel, and multiple electromagnetic signals are acquired in real time, and the electromagnetic signal is expanded, the cold-rolled thin strip steel plastic strain ratio on-line detection method is established by artificial neuron network algorithm model, realizes the purpose of on-line accurate measurement strip steel plastic strain ratio, scientific performance is strong, practical performance is high, reduces the waste of raw material and manpower cost.

This invention provides an intelligent optimization method and device for an atmospheric distillation unit. The method includes: collecting correlation parameters associated with key parameters of the atmospheric distillation unit, wherein the correlation parameters originate from on-site instrument measurement and analysis data of the atmospheric distillation unit and a process flow simulation model of the atmospheric distillation unit; inputting the correlation parameters into a pre-established artificial neural network mathematical model of the atmospheric distillation unit to obtain the key parameters of the atmospheric distillation unit; and providing a real-time optimization scheme design based on the predicted key parameters and the real-time operating status of the atmospheric distillation unit. This invention utilizes an artificial neural network model of the atmospheric distillation unit to predict and optimize key parameters that are difficult to measure and analyze in real time, thereby controlling the overall tower operation state to near-optimal levels, balancing product quality and energy conservation.

The invention provides a wide-area refresh rate display driving system based on a neuromorphic device and a working method of the wide-area refresh rate display driving system, and belongs to the technical field of display. The control module comprises an enabling control module and a brightness control module which are connected with each other, the input module is connected with the enabling control module, and the light-emitting module is connected with the brightness control module; the input module is used for splicing an input enable signal and an input data signal and inputting the signals into the control module; the enabling control module adopts artificial nerve components, and each artificial nerve component is switched between an on state and an off state according to an input enabling signal; the brightness control module adopts an artificial synaptic device, and a single artificial synaptic device is converted among a plurality of different conductivity states according to an input data signal; and the light-emitting module carries out light-emitting display under the regulation and control of the brightness control module. According to the invention, pixel-level control can be realized through artificial neurons, and wide-area refresh rate can be realized through artificial synapses.

The embodiment of the invention discloses a bridging artificial neuron unit based on a memristor circuit and a bionic auditory positioning system, and relates to the field of neuromorphic calculation and artificial intelligence hardware. A bridging artificial neuron unit comprises a first memristor analog circuit connected with a first capacitor and a first load resistor and a second memristor analog circuit connected with a second capacitor and a second load resistor, and the first memristor analog circuit and the second memristor analog circuit are connected through a first bridging capacitor and are coupled with an external circuit through a second bridging capacitor. The unit can simulate pulse distribution and non-stress behaviors of biological neurons. A plurality of units are connected in series to construct a one-way pulse propagation chain to form a pulse propagation network. The coincidence detector constructed based on the network can detect the tiny time difference of multiple paths of input signals, and is applied to a bionic auditory system to realize sound source positioning. The device is simple in structure, high in modularization degree and capable of achieving high-reliability analog pulse propagation and high-time-precision coincidence detection.

A device, system, and method is provided for training or prediction using a cluster-connected neural network. The cluster-connected neural network may be divided into a plurality of clusters of artificial neurons connected by weights or convolutional channels connected by convolutional filters. Within each cluster is a locally dense sub-network of intra-cluster weights or filters with a majority of pairs of neurons or channels connected by intra-cluster weights or filters that are co-activated together as an activation block during training or prediction. Outside each cluster is a globally sparse network of inter-cluster weights or filters with a minority of pairs of neurons or channels separated by a cluster border across different clusters connected by inter-cluster weights or filters. Training or predicting is performed using the cluster-connected neural network.

The application discloses an artificial neuron, comprising a microwave unit, a device unit and a weight unit. The microwave unit is used for generating a microwave signal; the device unit is responsive to the microwave signal and outputs a direct current voltage signal conforming to a plurality of clock curves with different peak response points; and the weight unit assigns a weight value to adjust the strength of the direct current voltage signal. The artificial neuron based on group coding with high flexibility can realize a single neuron with a plurality of adjustable nonlinear activation characteristics and can adapt to the needs of different artificial intelligence computing scenarios.

A semiconductor device that implements artificial neurons and synapses together on the semiconductor device includes a plurality of fins formed on the semiconductor device and a plurality of gates formed around the plurality of fins to form a plurality of fin field effect transistors (FinFETs). The plurality of FinFETs can form one or more artificial synapses and one or more artificial neurons. Each of the one or more artificial synapses can include two or more of the plurality of gates. Each of the one or more artificial neurons includes one of the plurality of gates.