43 results about "Reservoir computing" patented technology

The invention relates to an AI processor and method based on storage and calculation integration, three-dimensional integration and operator separation. The AI processor comprises a storage layer; the calculation layer and the storage layer are stacked in the vertical direction through a three-dimensional integrated connection structure, and a three-dimensional integrated high-speed channel is generated and used for executing calculation tasks in a large language model; the standardized memory interface is used for being connected with an external main control chip; the calculation task comprises a pre-filling stage and a decoding stage; the scheduling module configures the main control chip to interact data with the storage layer through the standardized memory interface so as to execute the pre-filling stage; the scheduling module configures the computing layer to interact data with the storage layer through the three-dimensional integrated high-speed channel so as to execute the decoding stage, the pre-filling stage is processed by utilizing the large computing power advantage of a main control chip, and the decoding stage is processed by utilizing the three-dimensional stacked high-bandwidth advantage, so that accurate matching of the computing power and the bandwidth is realized; and the large model reasoning efficiency is obviously improved.

The invention belongs to the technical field of reservoir calculation, and relates to a multi-dimensional pulse reservoir calculation method and system based on a semiconductor laser. The multi-dimensional pulse reservoir calculation method comprises the following steps: encoding a signal to be processed to generate an original optical signal; injecting the original optical signal into a semiconductor laser, and collecting a spike pulse optical signal containing to-be-processed signal feature information generated by the semiconductor laser under excitation of the original optical signal; performing photoelectric conversion on the spike pulse optical signal to obtain a time domain signal; amplitude features, quantity features and time position features of spike pulses in the time domain signals are extracted, and amplitude feature virtual nodes, quantity feature virtual nodes and time position feature virtual nodes are constructed; splicing the amplitude feature virtual node, the quantity feature virtual node and the time position feature virtual node to obtain a reserve pool state vector of an extended dimension; and obtaining an output signal of the multi-dimensional pulse reservoir calculation system based on the reservoir state vector of the extended dimension.

The invention belongs to the technical field of Internet of Things. The invention provides a large-scale infrastructure network link dynamic prediction method based on nonlinear reserve pool calculation. According to the embodiment of the invention, by combining nonlinear reserve pool calculation and regularization regression technologies, the dynamic, efficient and accurate prediction of a complex network link is realized, and the method is particularly suitable for state inference and stability analysis of a key link in an infrastructure network. A reserve pool calculation framework is adopted, and the core advantage of the method is that only the linear weight of an output layer needs to be trained, and the internal connection of a huge reserve pool is kept random and fixed. A large amount of iterative computation and GPU resources required by back propagation are avoided, and the training complexity and the time cost are remarkably reduced. A Tikhonov regularization item is introduced, so that the overfitting problem under limited training data or noise interference is effectively prevented, and the generalization ability and prediction stability of the model are improved.

The invention discloses a time delay reserve pool calculation method based on dispersion orthogonal polarized light feedback VCSEL. An adopted time delay RC system comprises a tunable laser, a Mach-Zehnder modulator, a VCSEL, an optical isolator, an optical coupler, a polarization controller, a variable optical attenuator, a chirp fiber Bragg grating and a photoelectric detector. A signal of the input layer is modulated by a Mach-Zehnder modulator and then is injected into an X polarization component of the VCSEL of the reserve pool layer; emitted light of the VCSEL is divided into two paths after passing through the optical circulator, one path is reflected back to the VCSEL by the chirp fiber bragg grating after being subjected to cross-polarization rotation through the polarization controller, and a reserve pool layer of the time delay RC system is constructed; one path is transmitted to the output layer and is converted into an electric signal through a photoelectric detector. According to the method, the time delay characteristic can be further inhibited, so that generation of harmful resonance is further inhibited, and the method has stronger nonlinear characteristic and better calculation performance.

This associative memory device comprises: a classification vector input data setting unit (1) that sets the same classification vector in a fixed time length as a first input vector; a time-series signal output data setting unit (2) that sets a vector composed of values at each time of a time-series signal in the time length as a first output vector; a first reservoir learning / calculation unit (3) that performs learning by reservoir computing in which a specific classification vector among a plurality of different classification vectors is associated with a specific time-series signal among a plurality of different time-series signals such that the classification vector becomes an input and the time-series signal becomes an output, and calculates a first recursive neural circuit that associates the first input vector with the first output vector; and a first reservoir circuit storage unit (4) that stores the first recursive neural circuit.

In reservoir computing, this enables calculations based on more past input information. [Solution] An information processing device that trains a model to estimate the correct output based on input has a model that includes three layers: an input layer, an intermediate layer that outputs an output value corresponding to the input value input from the input layer based on its internal state, and an output layer that outputs a value based on the output value output from the intermediate layer. In the intermediate layer, a reservoir layer, which is a collection of multiple nonlinear nodes that each perform nonlinear operations, and a linear node layer, which is a collection of multiple linear nodes that each perform linear operations, are connected in parallel to the input layer.

The application belongs to the technical field of reservoir computing, and relates to a multi-dimensional pulse reservoir computing method and system based on a semiconductor laser. The multi-dimensional pulse reservoir computing method comprises the following steps: encoding a to-be-processed signal to generate an original optical signal; injecting the original optical signal into a semiconductor laser, and collecting a spike pulse optical signal containing characteristic information of the to-be-processed signal generated by the semiconductor laser under the excitation of the original optical signal; photoelectrically converting the spike pulse optical signal to obtain a time-domain signal; extracting amplitude characteristics, quantity characteristics and time position characteristics of the spike pulse in the time-domain signal, and constructing amplitude characteristic virtual nodes, quantity characteristic virtual nodes and time position characteristic virtual nodes; splicing the amplitude characteristic virtual nodes, the quantity characteristic virtual nodes and the time position characteristic virtual nodes to obtain an extended-dimension reservoir state vector; and obtaining an output signal of the multi-dimensional pulse reservoir computing system based on the extended-dimension reservoir state vector.

The invention provides a reservoir calculation method based on rare earth doped nanocrystals. The reservoir calculation method comprises the following steps: acquiring an original digital image and converting the original digital image into a binary vector set; modulating the binary vector set into a pulse sequence set through time multiplexing coding; inputting the pulse sequence into a reservoir formed by three layers of core-shell-shell structure rare earth doped nanocrystals, and generating a corresponding luminescence response curve set through nonlinear dynamic interaction of the three layers of core-shell-shell structure rare earth doped nanocrystals; extracting features based on the curve set to form a reservoir state vector set; and finally, training and classifying the state vector set through a linear layer to obtain an image classification result. According to the reservoir calculation method based on the rare earth doped nanocrystals, through the synergistic effect of the physical characteristics of the rare earth doped nanocrystals and time multiplexing coding, a single reservoir replaces a traditional complex electronic network, coding, calculation and feature extraction are completed in an optical domain, the algorithm complexity and the training cost are remarkably reduced, and the calculation efficiency is improved. And efficient image classification is realized.

The application relates to a discrete dislocation dynamics acceleration simulation method based on reservoir computing, which effectively reduces the high computing cost caused by the global time step limit and the long-range interaction of dislocation elastic force in traditional dislocation dynamics simulation by replacing the node motion speed calculation in discrete dislocation dynamics with a trained reservoir computing model. The virtual node mechanism is innovatively proposed to solve the dimension mismatch problem caused by the dynamic change of the number of nodes in the dislocation topology evolution process, and adaptive discretization prediction is realized. The prediction error accumulation is prevented by regularly inserting the real traditional dislocation dynamics simulation calculation result in the prediction sequence, and the stability of long-term simulation is ensured. The acceleration scheme significantly reduces the computing time complexity from exponential to linear while maintaining the physical accuracy, providing a feasible technical path for efficient simulation of large-scale three-dimensional complex dislocation network evolution.

The invention relates to a database operation method and device for function calculation, computer equipment, a computer readable storage medium and a computer program product. The method applied to any function computing node in a function computing cluster comprises the steps that a computing layer in a current function computing node receives a function computing request, wherein the function computing request comprises an object identifier and a function identifier; if the data of the object identifier is stored in the storage layer of the other function computing node, the computing layer forwards the function computing request to the other function computing node; if the data is stored in the storage layer of the current function calculation node, the calculation layer starts the database transaction, the function instance is executed based on the data of the object identifier, and when the function is executed, reading operation is performed on the distributed storage database provided by the storage layer through the database transaction, and data required by writing operation on the distributed storage database is cached; and after the function execution is finished, the database transaction is closed, the cached data is submitted to a storage layer for write operation, and the communication overhead is reduced.

The application belongs to the technical field of neuromorphic computing devices, and particularly relates to a temperature programmable neuromorphic device based on lead sulfide colloidal quantum dots and application. The device comprises an insulating substrate, a planar gap electrode, a lead sulfide colloidal quantum dot photosensitive film and a temperature control module. By setting the working temperature to control the release kinetics of the trap state charge in the photosensitive film, the device has different photocurrent decay characteristics at different temperatures, realizing reversible switching of the fast detection and long-term storage modes. The device can be used for reservoir computing, and realizes time sequence information processing through the historical dependence response to time sequence light pulses. The application solves the problem of fixed and unadjustable response time scale of the existing device, and has the advantages of simple structure, good process compatibility and wide response dynamic range.

The invention relates to a discrete dislocation dynamics acceleration simulation method based on reservoir calculation, and the method comprises the steps: replacing the node movement speed calculation in discrete dislocation dynamics with a trained reservoir calculation model; the high calculation cost caused by global time step length limitation and dislocation elastic force long-range interaction in traditional dislocation dynamics simulation is effectively reduced. A virtual node mechanism is innovatively provided, the problem of dimension mismatching caused by dynamic change of the number of nodes in the dislocation topology evolution process is solved, and self-adaptive discretization prediction is achieved. And real traditional dislocation dynamics simulation calculation results are periodically inserted into the prediction sequence, so that prediction error accumulation is prevented, and the stability of long-term simulation is ensured. According to the acceleration scheme, while the physical accuracy is kept, the calculation is obviously reduced from the index time complexity to the linear time complexity, and a feasible technical path is provided for efficient simulation of large-scale three-dimensional complex dislocation network evolution.

The invention discloses a method for realizing reserve pool calculation based on organoid, which comprises the following steps: acquiring an audio signal, and converting the acquired audio signal into a photoelectric stimulation pulse signal; acquiring a fluorescence signal responding to the photoelectric stimulation pulse signal through the organ-like device; according to the obtained fluorescence signals, high-dimensional feature data are obtained through calculation by using a storage pool; and decoding the high-dimensional feature data to complete identification processing of the audio signal. According to the method, a biological organoid is adopted as a core of a reserve pool, energy consumption is extremely low when the organoid executes perception and cognition tasks depending on the natural storage and calculation fusion characteristic of a biological neural network, organic fusion of data storage and processing is achieved through neural synapses, a large amount of energy consumption of CPU and memory data transmission in a traditional von Neumann architecture is avoided, and the reliability of the von Neumann architecture is improved. And the development requirement of low-power AI hardware is better met, and a thought is provided for developing an efficient and low-cost neuromorphic chip.

The invention particularly relates to a physical reservoir for a magnetic reservoir computing device. The physical reservoir includes a ferromagnetic film including a two-dimensional arrangement of point deformations. The point deformations are dimensioned to function as pinning sites of magnetic domains of the ferromagnetic film. The invention also relates to a magnetic reservoir computing device comprising such a physical reservoir, and methods of operating and manufacturing such a reservoir computing device. The proposed method results in a low power consumption physical reserve that is easy to manufacture.

Disclosed is a synaptic device, a reservoir computing device using the synaptic device, and a reservoir computing method using the reservoir computing device. The synaptic device includes a substrate and a plurality of units cells on the substrate, wherein the unit cells each include a channel layer and a first electrode and second electrode intersecting the channel layer, wherein the first electrode and the second electrode are spaced apart from each other, and define a gap region exposing a portion of the channel layer, and the channel layer includes a 2-dimensional semiconductor material or a 2-dimensional ferroelectric material.

This invention provides a depth-scalable inversion method and system for 3D ground-penetrating radar (GPR) data. The method includes: spatially mapping the 3D GPR volume data of the area to be inverted point by point according to spatial coordinate points based on a 3D reservoir computing network to obtain waveform dynamic features characterizing the electromagnetic response variation law of the underground medium; extracting global spatial features of the 3D GPR volume data based on a convolutional neural network; predicting the target category of the 3D GPR volume data based on the waveform dynamic features and global spatial features; determining the inversion network depth based on the target category; constructing an encoder-decoder inversion network with corresponding layers based on the inversion network depth; using the encoder-decoder inversion network to extract multi-scale spatial features and texture structure features of the 3D GPR volume data; and predicting the 3D distribution of the underground dielectric constant corresponding to the 3D GPR volume data based on the multi-scale spatial features and texture structure features.

The present application belongs to the technical field of reservoir computing, and particularly relates to a method for implementing a large-scale optoelectronic reservoir computing system. The method comprises: performing matrix sparsification connection to form a sparse connection matrix in the form of a lower triangular matrix composed of basic matrix calculation units A and B; then performing rank reduction operation on the sparse connection matrix, so that the product of the original sparse matrix and the input signal is converted into the product form of each basic matrix unit after splitting and the input signal after splitting, and the dimension of each subunit is reduced to 1 / 2 of the original dimension; the above splitting operation is repeated until the splitting reaches the scale supported by the optical chip unit; then the reservoir is trained and tested to obtain a weight matrix calculation prediction value. The multiplication operation shares the same chip, and the multiplication and accumulation operation of a matrix of any scale can be completed by multiplexing the basic scale optical computing chip. The present application achieves ideal effects in the application of communication signal post-equalization, signal recognition, etc.

A reservoir computing system comprises an input system configured to generate an input signal, an optical reservoir configured to receive the input signal from the input system, and an output system configured to generate an output signal based on light received from the optical reservoir. The optical reservoir comprises a plurality of reservoir lasers configured for injection-locked operation and a reservoir randomization element configured to randomly distribute light output by the plurality of reservoir lasers among the plurality of reservoir lasers to contribute to injection locking of the plurality of reservoir lasers.

The invention relates to the technical field of neuromorphic calculation and brain-like information processing, in particular to a liquid reservoir calculation system with distributed and globally adjustable dynamics, which comprises a liquid storage module, a plurality of reservoir nodes, an input signal application module, an output signal acquisition module, a liquid state regulation and control module and a readout layer module. A plurality of reserve pool nodes are simultaneously located in the same electrolyte liquid environment, the input signal applying module applies a time change electric signal to excite each node to generate a dynamic response, and the output signal acquisition module acquires a dynamic response signal of each node. Because the dynamic response time constants of the nodes are different, the system forms distributed dynamics with multi-time scale characteristics. And the liquid state regulation and control module realizes the adjustability and reconstruction of the overall dynamic behavior of the system under the condition of not changing the physical structure of the node by regulating and controlling the overall parameters of the electrolyte liquid. And the readout layer module processes the dynamic response signal and is used for tasks such as time sequence prediction, pattern recognition or dynamic signal modeling.

The application discloses a kind of photonic reservoir computing system based on reflective semiconductor optical amplifier, comprising: input layer, reservoir layer and output layer;Wherein, the pre-processing of signal is carried out in input layer, the masking process to input signal is completed, and the masked input signal is injected into reservoir layer;In the reservoir layer, the output light of RL is sampled at equal time interval, and N sampling points are obtained as the virtual nodes of reservoir;In the output layer, the state of all virtual nodes from RL to RSOA delay line is recorded, and the optimal readout weight is calculated by ridge regression algorithm, then the virtual node state and readout weight are linearly weighted and summed, to obtain the final result.

The invention discloses a transient stability assessment method and system for a power system. The method comprises the following steps: firstly, acquiring multi-dimensional real-time electric quantity data of a power system; thirdly, performing feature selection optimization by applying a knowledge-guided memetic algorithm, combining global evolutionary search and local fine search and fusing expert knowledge to obtain an optimal feature combination and a time window; then, constructing a reservoir calculation network formed by differential neuron ring oscillators based on the combination, and extracting deep dynamic features of the system; and finally, inputting the deep features into a CNN-LSTM hybrid model for evaluation, and obtaining a transient stability probability value. According to the method, through innovative feature optimization and dynamic feature extraction mechanisms and in combination with the deep fusion model, rapid and accurate evaluation of the transient stability of the power system is realized, and the safety protection level of a power grid is remarkably improved.

The invention discloses an electroencephalogram signal bad track detection and reconstruction method and device based on deep learning and reservoir calculation fusion, and the method comprises the steps: carrying out the segmentation processing of an obtained multi-channel electroencephalogram signal, and dividing the data of each channel into data segments of a fixed time window; inputting the data segments into a BC-LSTM bad track detection module for abnormal segment identification, and classifying the channels as normal channels or bad tracks according to the proportion and distribution characteristics of the abnormal segments; for a bad track, when enough normal data segments exist, using the data to train a CWRC model to perform signal reconstruction; when normal data segments are scarce or completely missing, adopting a transfer learning method driven by neural electrophysiology similarity to perform reconstruction; and integrating the reconstructed bad track signal with an original normal signal, and outputting complete electroencephalogram data. According to the method, the full-automatic electroencephalogram bad track processing flow is achieved, and a complete solution is provided for large-scale clinical electroencephalogram preprocessing and neuroscience research and application.

A reservoir computing method (700) and related system (100) are disclosed. The reservoir computing system includes at least one non-linear computing node (103) with a computing node MOSFET (113), a drive circuit (110) coupled with the at least one non-linear computing node (103), and a readout unit (104) for acquiring an output sequence associated with the at least one non-linear computing node (103). The MOSFET has charge trapping sites positioned in a gate dielectric that exhibit a bias temperature instability when subjected to a voltage stress and exhibit a threshold voltage recovery after removal of the voltage stress. The drive circuit is configured to apply an input pulse and a readout pulse to a gate of the MOSFET. The readout unit (104) is configured to detect a current flowing through a channel region of the MOSFET when the drive circuit is applying the readout pulse. The readout pulse has a smaller voltage amplitude than the input pulse.

The application relates to a memristor reserve pool computing circuit and a prediction method for real-time epilepsy prediction, and belongs to the technical field of electroencephalogram signal processing. The circuit comprises a plurality of levels of series-connected reserve pool computing architectures. Each level of architecture is used for receiving a multichannel EEG signal or a multichannel binary coded signal, and after high-dimensional feature mapping, nonlinear time feature extraction and feature fusion are sequentially performed on the multichannel EEG signal or the multichannel binary coded signal by a mask circuit, a reserve pool module based on a volatile memristor and a multiply-accumulate circuit comprising a cross array based on a non-volatile memristor in the architecture, the parallel output road fusion signal is output to a feature binaryzation circuit in the architecture, signal binaryzation processing is performed on the parallel output road fusion signal, a multichannel binary coded signal is output in parallel as an input of a next level of architecture, and finally, a binary coded signal output by a last level of architecture is output as an epilepsy prediction result. The application provides a hardware efficient solution for real-time low-power epilepsy seizure prediction.

The application discloses a distributed optical fiber sensing system based on delay reservoir computing and a signal identification and positioning method thereof. The system is composed of an input layer, a reservoir and an output layer. The input layer has two nodes, one of which inputs a non-essential reservoir excitation signal and the other of which inputs an external vibration signal; the reservoir is composed of a delay nonlinear system, and a delay optical fiber thereof is a sensing optical fiber; the output layer has two nodes, one of which outputs a vibration type and the other of which outputs a vibration position. The signal identification and positioning method comprises the following steps: before the system is started, network training is performed, and output connection weights are determined through target outputs and virtual node states of training signals; after the system is started, the output connection weights and virtual node states of a vibration signal to be measured are used to obtain an output of the reservoir calculation, i.e. a vibration signal type and a vibration position. The application realizes accurate and real-time identification and positioning of external vibrations by using the powerful learning ability, modeling ability and fast processing ability of optical reservoir calculation.

The invention relates to the field of photoelectric devices and neuromorphs, in particular to a reserve pool computing device based on a heterojunction and a preparation method of the heterojunction. The reservoir computing device comprises a plurality of heterojunctions, and each heterojunction comprises a substrate, a p-type GaN layer arranged on the substrate, a GaOx layer covering the surface of the p-type GaN layer, a single-layer MoS2 covering the surface of the GaOx layer, a first metal electrode in contact with the p-type GaN layer and a second metal electrode in contact with the single-layer MoS2. By adopting the method, stable synaptic response can be generated.

A reservoir computing method and related system. The method comprises providing a pulse wave representation of a reservoir input and a MOSFET-based compute node of a reservoir computing system. A body terminal of the MOSFET is configured to apply a bias signal to the body region. The MOSFET is operable in floating mode in the absence of the bias signal. The MOSFET displays a history-dependent threshold voltage when driven by input pulses in floating mode. The method further includes removing the bias signal from the body terminal; applying input pulses to a gate terminal of the MOSFET, thus causing history-dependent variations in the threshold voltage; and acquiring an output sequence related to the compute node while the MOSFET threshold voltage is varying, by repeatedly applying readout pulses to the gate terminal of the MOSFET and simultaneously detecting a current through a channel region of the MOSFET.

This invention discloses a room-temperature mode-selective physical reservoir computation method, belonging to the field of neuromorphic computing. The method is based on an artificially synthesized antiferromagnetic skyrmion spring oscillator, which electrically selects different intrinsic dynamic modes to adapt to different computational tasks: an in-plane alternating current drives a large-amplitude oscillation mode, providing strong nonlinearity to optimize signal conversion tasks; an out-of-plane alternating magnetic field drives a breathing mode, providing strong short-term memory to optimize time series prediction tasks. The device integrates fully electrically driven, multi-channel parallel readout, voltage-controlled magnetic anisotropy stabilization, and planar microcoil excitation units, achieving full on-chip integration of mode selection, signal processing, and stability control. This invention solves the problems of nonlinearity and memory coupling, poor task adaptability, and insufficient room-temperature stability in physical reservoirs, achieving fast, high-precision, and reconfigurable processing of multiple tasks at room temperature using a single device.

The invention discloses a reconfigurable neuromorphic olfactory sensing memristor system and a preparation method thereof, and relates to the technical field of neuromorphic electronics and intelligent sensing, and the system comprises a sensing calculation module which comprises a memristor based on an MXene (at) SnS2 (at) PANI heterostructure, the memristor shows a synaptic characteristic under gas stimulation and shows a neuron characteristic under electric pulse stimulation, and the memristor shows a neuron characteristic under electric pulse stimulation; a sensing signal is output; the signal conditioning and encoding module receives the sensing signal and encodes the sensing signal into a pulse sequence; the neuromorphic calculation module receives the pulse sequence and executes gas concentration identification and gas flow rate identification according to the reservoir calculation network and the pulse neural network; and the decision and execution module generates a control instruction according to an output result of the neuromorphic calculation module so as to trigger a corresponding decision operation. Integrated integration of gas sensing and neuromorphic calculation is realized, and high-efficiency and low-power-consumption intelligent gas monitoring is realized.