1833 results about "Neuron" patented technology

The invention provides an end-side multi-modal large model accelerated reasoning method and system, and the method comprises the steps: carrying out the two-stage screening and rearrangement of visual tokens based on the CLS attention and text-to-visual attention in a visual encoder and pre-filling stage, and constructing a sparse attention and sparse key value cache; in a decoding stage, an important neuron set is judged according to activation gating or historical statistics, only a corresponding feedforward network weight is pulled and calculated, missed weights are loaded on demand through asynchronous I / O, and hot neurons are maintained in a high-speed memory to utilize model sparsity, so that video memory / memory occupancy and calculation overhead are remarkably reduced on an end side; throughput and time delay performance are improved. According to the method, the internal memory and computing resources required by reasoning of the multi-modal large language model are reduced from two dimensions by utilizing the endogenous sparsity of the end-side large language model in input and the model, so that a higher reasoning speed is achieved by utilizing fewer resources on the premise of keeping the size of the model unchanged, and the performance of the whole system is improved.

The invention provides a transcranial stimulation magnetic therapy closed-loop control method and system, and the method comprises the steps: determining a multi-modal physiological signal of a user, the multi-modal physiological signal at least comprising an electroencephalogram signal, an autonomic nerve physiological signal and a subjective feedback signal; constructing an individual dynamic baseline spectrum based on user multi-state data collected before treatment; the method comprises the following steps: preprocessing a multi-modal physiological signal collected in a treatment process, and extracting a real-time feature vector; calculating a comprehensive deviation index of the real-time feature vector and the individual dynamic baseline spectrum, and determining a deviation type based on the comprehensive deviation index; the adjustment amount of transcranial magnetic stimulation parameters is dynamically calculated according to the deviation type, the individual response characteristics and the historical treatment data, stimulation operation is conducted according to the adjustment amount of the transcranial magnetic stimulation parameters, and the transcranial magnetic stimulation parameters at least comprise the stimulation frequency, the intensity, the pulse width and the stimulation duration. Parameters of transcranial magnetic stimulation are adaptively changed according to changes of neuron oscillation of a patient in the treatment process.

The invention discloses a closed-loop noninvasive spinal cord electrical stimulation regulation and control method and system based on motion data, and the system is provided with a flexible electrode array patch which is attached to the skin surface of a to-be-tested human body; the sensing module is used for collecting electromyographic signals or acceleration signals of a user; the signals are processed, and characteristic values related to actions and motions are extracted; the closed-loop feedback module compares the characteristic value with a model set characteristic / threshold value, and judges whether to start stimulation site and stimulation parameter adjustment or not; automatically selecting an optimal electrode unit and a corresponding stimulation parameter based on a judgment result; the dynamic regulation and control module controls the optimal electrode unit to output stimulation current, and regulates the activity of corresponding spinal cord neurons; and continuously performing closed-loop adjustment on the electrode unit and the stimulation parameters according to the motion feedback data. According to the method and system, precise stimulation and closed-loop regulation and control based on the motion response of the patient are achieved, and the problems that stimulation target positioning depends on artificial experience, parameter adjustment is not timely, and individualized and closed-loop feedback capacity is lacked are solved.

The invention is applied to the field of mechanical fault diagnosis signal processing, and particularly provides a pulse neural network mechanical fault diagnosis method based on STFT dimension transformation, and the method comprises the steps: collecting a one-dimensional mechanical vibration signal, carrying out the wavelet decomposition, carrying out the wavelet reconstruction of a low-frequency component and a denoised high-frequency component, and carrying out the wavelet reconstruction of the low-frequency component and the denoised high-frequency component; obtaining a denoised one-dimensional vibration signal; performing short-time Fourier transform, and converting the time-frequency two-dimensional matrix into a time-frequency two-dimensional matrix; inputting the time-frequency two-dimensional matrix into an improved HH threshold neuron model, carrying out Poisson sparse coding on the time-frequency two-dimensional matrix, and only carrying out pulse response on signal significant features; constructing a suprathreshold coding convolutional network with residual connection, inputting a sparse coding matrix, training by adopting an unsupervised learning rule based on STDP, and adaptively adjusting a network synaptic weight; and inputting to a trained above-threshold coding convolutional network, and obtaining pulse emission activity of neurons of an output layer through network forward propagation to determine a fault diagnosis result.

The invention belongs to the field of bioinformatics, and relates to an improved integrated deep learning cell communication ligand-receptor interaction prediction method. The method comprises the following steps: firstly, carrying out extraction and dimensionality reduction on biological sequence features of a ligand and a receptor, and constructing multi-modal feature input; secondly, constructing an improved deep neural network branch, introducing a batch normalization layer and a Leaky ReLU activation function, solving the problems of gradient disappearance and neuronal necrosis, and improving regularization strength to prevent overfitting; meanwhile, an enhanced heterogeneous graph auto-encoder branch is constructed, the graph embedding dimension is remarkably expanded to improve the feature capacity, and full convergence of the model is ensured by increasing training rounds; thirdly, fusing the improved deep network with the prediction probability of a heterogeneous graph auto-encoder by adopting a weighted integration strategy; and finally, outputting a potential interaction relationship based on the fusion probability. By optimizing the architecture and the strategy, the prediction accuracy and robustness are remarkably improved, and a reliable tool is provided for analyzing a complex cell communication network.

The invention discloses a pulse neural network hardware accelerator and a data processing method, and the accelerator is characterized in that a low-power-consumption three-stage pipeline CPU module is used for receiving input data, scheduling an SNN network acceleration instruction, and sending the input data to an asynchronous edge SNN hardware accelerator module through a coprocessor interface; the asynchronous edge SNN hardware accelerator module comprises a pulse data encoding and decoding module, L neuromorphic kernels and an on-chip network, the pulse data encoding and decoding module encodes input data into a pulse form and sends the pulse form into the neuromorphic kernels, and the neuromorphic kernels are used for performing calculation based on the data in the pulse form; the network-on-chip is used for communication between the neuromorphic kernels, and the connection between neurons before and after synapses in the neuromorphic kernels is realized by adopting a synaptic cross array. According to the invention, the data processing acceleration performance can be greatly improved.

The invention provides a satellite weak and small target on-orbit observation system and method based on intelligent closed-loop feedback, and the system comprises a detection module which carries out the image enhancement and time sequence consistency enhancement of an infrared image, obtains an enhancement feature, and obtains a candidate target set based on the enhancement feature; the tracking module is used for carrying out target matching in the search area; when the matching succeeds, the candidate position is used as an observation value, and the observation value is input into a Kalman filter to obtain a target state vector; when the matching fails, taking a prediction state of the Kalman filter as a target state vector, expanding a search window by taking a prediction position as a center, and executing re-identification; the control module is used for mapping the target state vector into an attitude error and generating a control instruction by adopting a single-neuron self-adaptive PID (Proportion Integration Differentiation) controller; and the closed-loop scheduling module is used for dynamically adjusting operation parameters of at least one module according to the execution error and the detection confidence coefficient. According to the invention, the continuous tracking precision and attitude control stability of the weak and small target are improved.

The invention relates to a human body health state assessment system and assessment method based on multi-point acquisition, and the system comprises an event sensing module which is used for collecting an image data signal, a physiological signal and an environment data signal of a human body, and outputting an asynchronous pulse event flow when the signal change is detected; the pulse coding module is used for generating a space-time pulse sequence through a preset adaptive threshold coding technology; the neural network processing module is used for performing event driving processing by utilizing leakage integral distribution neurons in a preset pulse neural network according to the space-time pulse sequence to obtain a current health characteristic pulse mode; and the health state decoding module is used for obtaining a health state score and a risk classification result of the current human body through a preset pulse distribution rate analysis algorithm and a preset time sequence decoding technology according to the current health characteristic pulse mode. Therefore, the problems of high power consumption, high delay, low data transmission efficiency, low early pathological feature recognition accuracy and the like of a human health state evaluation system are solved.

The invention belongs to the technical field of machine state prediction, and particularly relates to a fault diagnosis method and system based on a coding and decoding attention mechanism enhanced pulse neural network, and the method comprises the steps: collecting a device vibration signal, and segmenting the device vibration signal into time sequence segments; converting each fragment into a two-dimensional time-frequency image by using S transformation to construct a data set; the image is input into a lightweight hybrid network model to extract deep features, the model is composed of a pulse convolution encoder and a pulse efficient additive attention mechanism encoder which are alternately cascaded, and the pulse convolution encoder extracts local features through pulse neurons and depth separable convolution; the global context dependency is modeled by adopting an efficient additive attention mechanism with low linear complexity; and finally, fault classification is completed based on deep features. Through fusion of pulse calculation and an efficient attention mechanism, the feature extraction capability and diagnosis precision are improved while the parameter quantity of the model is remarkably reduced, and the method is particularly suitable for being deployed on edge equipment with limited resources to realize fault diagnosis.

The invention discloses an environment-emotion gradient association memristive neural network circuit based on a TAP cell regulation mechanism. The environment-emotion gradient association memristive neural network circuit is composed of a hippocampus module and an amygdala kernel module. The hippocampus module comprises an environment recognition neuron module and an emotion generation neuron module; the environment recognition neuron module and the emotion generation neuron module are each composed of a weight regulation and control module adj based on a TAP cell regulation mechanism, an adder, a memristor, a resistor, an operational amplifier, a function module ABM1 and a sampling holder. Wherein the weight regulation and control module adj is composed of a voltage-controlled switch, a summator, a function module ABM2 and a function module ABM3. The almond kernel module is composed of a weight regulation and control module adj, a memristor, a resistor, an operational amplifier and a sampling holder. The provided circuit simulates the hippocampus and amygdala nucleus of the brain in structure, the regulation principle of TAP cells on neural stem cells is used for reference in a weight regulation mechanism, and bidirectional gradient associated memory of the environment and the emotion can be achieved.

The invention discloses a model security test method, device and equipment and a storage medium, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining a test sample composed of a test image and a corresponding target detection label, and a to-be-tested visual target detection model used for executing a target detection task, and generating an adversarial sample based on the test sample by using the to-be-tested visual target detection model, respectively inputting the test sample and the adversarial sample into the to-be-tested visual target detection model to obtain a corresponding prediction result so as to determine a target test score of a target neuron in the to-be-tested visual target detection model, and generating a security test result. By comparing and quantizing the neuron fraction through the results of the adversarial sample and the test sample, the sensitive neuron in the model is determined, the model safety can be determined more accurately, and the anti-interference capability, safety and stability of the model are improved.

The invention discloses a wireless communication network intelligent optimization method and system based on neuron collaboration, and relates to the technical field of wireless communication. The method comprises the following steps: mapping a communication node into a bottom layer sensing neuron and constructing a neural state variable set; constructing a node neural situation function based on the variables; when the local threshold value is exceeded, excitation pulses are generated and uploaded to middle-layer convergence neurons; the middle layer carries out pulse space aggregation and extreme value search, and outputs a selection strategy and an adjustment strategy; and reporting to a top layer to execute whole network neural synchronization index analysis, and optimizing the wireless communication network. The technical problems of low spectrum resource utilization rate and unstable network performance caused by the fact that a traditional wireless communication network cannot realize high-efficiency spectrum allocation and dynamic topology reconstruction under user mobility change are solved, and the purposes of realizing local quick response and global collaborative optimization by constructing a layered neural network and improving the network performance are achieved. And the spectrum resource utilization rate is improved, and the network dynamic adaptive capability is enhanced, so that the user service quality is guaranteed.

The invention relates to the technical field of related drugs for nerve injury caused by heavy metal pollution, in particular to application of a compound for inhibiting phosphorylation of ARMC10 in preparation of drugs for antagonizing tin-induced nervous system injury. Exposure of trimethyltin chloride induces nerve cell mitochondrial dysfunction, and the key mechanism is ARMC10 serine 43 site abnormal phosphorylation. Phosphorylated protein causes excessive mitochondrial fission, membrane potential collapse and energy metabolism disorder, resulting in neuron damage. Based on the target spot, a compound for antagonizing tin-induced nervous system injury is obtained through screening, protein phosphorylation can be specifically inhibited, mitochondrial dynamic unbalance and dysfunction induced by trimethyltin chloride are effectively reversed, and neurotoxicity is relieved. According to the technical scheme, the technical problem that in the prior art, no medicine for effectively antagonizing tin-induced nervous system injury exists can be solved, and the compound has the application potential for treating the nervous system injury and cognitive impairment caused by tin exposure.

A method of facilitating a skill learning process or improving performance of a task, comprising: determining a brainwave pattern reflecting neuronal activity of a skilled subject while engaged in a respective skill or task; processing the determined brainwave pattern with at least one automated processor; and subjecting a subject training in the respective skill or task to brain entrainment by a stimulus selected from the group consisting of one or more of a sensory excitation, a peripheral excitation, a transcranial excitation, and a deep brain stimulation, dependent on the processed temporal pattern extracted from brainwaves reflecting neuronal activity of the skilled subject.

The disclosure relates to compositions and methods for treating a disease or condition associated with a TDP-pathology or a decline in TDP-43 functionality in neuronal cells in a subject, and for identifying candidate agents to restore expression of a normal full-length or protein coding STMN2 RNA.

The invention discloses a skin stretch forming residual stress rapid calculation method which comprises the following steps: randomly generating a combination of a pre-stretching amount, a coating elongation rate and a friction coefficient, submitting the combination to finite element analysis software to execute batch simulation, and generating a result file named by process parameters; traversing all the result files, extracting node numbers and residual stress values, and storing the node numbers and the residual stress values as a text format data set corresponding to the process parameters; a full-connection neural network model is constructed, an input layer receives the three process parameters of the pre-stretching amount, the coating elongation and the friction coefficient, a hidden layer comprises multiple layers of neurons and adopts a ReLU activation function, and an output layer generates residual stress values of all nodes; training the neural network model by using the data set, and adjusting the network weight through an optimizer; and inputting target process parameters to the trained neural network model, and outputting residual stress calculation results of all nodes of the skin. The technical purposes of rapidness, high efficiency and low cost are achieved.

The invention discloses a training method of a blood glucose control model and a blood glucose control method and system, and belongs to the technical field of blood glucose control. According to the method, a framework combining a blood glucose control model and a reinforcement learning model is constructed; wherein the blood glucose control model is a full-connection neural network with two layers; the number of neurons in the first layer of the network is 3, and an adopted activation function is a linear function; the number of neurons in the second layer of the network is 1, the adopted activation function is a nonlinear activation function, the output value domain of the nonlinear activation function is bounded, the upper bound is a positive number, and the lower bound is a negative number. The blood glucose control model is a model evolved after a classic PID algorithm is improved based on a blood glucose control task, the parameter space of the model comprises the parameter space of the classic PID algorithm, and the model has a wider range, can extract features with higher expression ability, is more adaptive to a reinforcement learning process, and can simply and efficiently realize accurate control of blood glucose.

The invention relates to an edge-guided cross-modal interactive fusion land cover classification method, which is particularly suitable for collaborative semantic segmentation of an optical image and a synthetic aperture radar (SAR) image. The method comprises the following steps: constructing a pseudo twin multi-stream encoder to respectively extract multi-scale features of optical and SAR images, embedding a cross-modal adaptive feature interaction module between encoding stages, realizing dynamic alignment and re-calibration of features between modals through a channel-level and space-level interaction mechanism, and relieving fusion deviation caused by heterogeneity and spatial dislocation. Neuron-level attention and multi-scale depth separable convolution are further fused through a lightweight feature fusion module, noise is suppressed, and context information is aggregated. And meanwhile, an edge auxiliary module is introduced to extract multi-scale boundary information from low-level features, and the multi-scale boundary information and semantic features are jointly decoded, so that the boundary detail and small target recognition capability is improved. According to the method, the precision and robustness of land coverage segmentation in a complex scene are remarkably improved on the premise of ensuring the calculation efficiency.

The invention provides a federated learning sub-model extraction method and system based on neuron dynamic perception, and belongs to the technical field of federated learning. A server initializes global model parameters and global model gradients; the client determines a model capacity proportion according to local hardware resources, a random strategy is adopted in the first round to extract a sub-model from the global model, and the neuron with the highest activity degree is selected according to the neuron activity degree in the subsequent round to form the sub-model; the client uses local data to train the sub-model and uploads the gradient to the server; and the server performs weighted averaging on the gradients uploaded by the clients, and updates global model parameters and global model gradients. According to the method, the sub-model extraction is guided through the neuron activeness, the neurons with slow convergence speed are preferentially trained, the global model performance is effectively improved, meanwhile, all the neurons can obtain training opportunities, and the method is suitable for a resource-limited edge device environment.

The invention provides a power grid constant value detection method and system, equipment and a storage medium, and belongs to the technical field of power grid detection, and the method comprises the steps: obtaining to-be-detected test constant value data; determining a target neuron matched with the test constant value data from the self-organizing mapping model, and determining a test error between the test constant value data and a weight vector of the target neuron; based on the test error and an error threshold value, determining whether the test constant value data is abnormal or not; wherein the error threshold value is determined by the following steps: acquiring each historical constant value data of the power grid; determining first neurons respectively matched with the historical constant value data from a self-organizing mapping model; and determining an error threshold value based on the weight vector corresponding to each first neuron and the historical constant value data corresponding to each first neuron. According to the power grid constant value detection method and system, the equipment and the storage medium provided by the invention, the recognition precision of the abnormal constant value can be improved.

Methods and compositions comprising one or more dopamine neuronal activity enhancers (e.g., dopamine receptor agonists), pantethine and solubilized curcumin for use in treating NAFLD and NASH are provided.Methods and compositions comprising one or more dopamine neuronal activity enhancers (e.g., dopamine receptor agonists), pantethine and solubilized curcumin for use in treating NAFLD and NASH are provided.

The invention discloses a composition for regulating sleep rhythm homeostasis as well as a preparation method and application thereof. The composition for regulating the sleep rhythm homeostasis is prepared from the following components in percentage by weight: 5 to 18 percent of polydextrose, 9 to 15 percent of galactooligosaccharide, 1 to 10 percent of inulin, 0.4 to 2 percent of gamma-aminobutyric acid, 0.8 to 2.2 percent of L-theanine, 0.1 to 3.5 percent of magnesium citrate, 2 to 21 percent of semen ziziphi spinosae extract, 1 to 10 percent of poria cocos extract, 0.5 to 2 percent of pericarpium citri reticulatae extract, 0.3 to 1 percent of liquorice root extract and 15 to 40 percent of maltodextrin. 5-18% of coconut oil powder MCTs, and 0.5-2% of a lily extract. According to the application, a magnesium MCTs shell-core structure is delivered by maltodextrin to form a mitochondrial piezoelectric engine so as to improve prebiotics, promote flora metabolism efficiency and energy supply feedback, activate intestinal pheochromophilic cells and improve SLC6A4 gene expression so as to improve synthesis of 5-hydroxytryptamine in the brain and activate GABA energy neurons; meanwhile, through the spina date seed and poria cocos quantum dots and a pericarpium citri reticulatae and liquorice pi-pi stacking network, the blood-brain barrier penetration rate is greatly increased, and hypothalamic CRH release is adjusted.

Disclosed are a neuron and a neuromorphic system including the same. More particularly, a neuron according to an embodiment of the present invention includes a completely depleted Silicon-On-Insulator (SOI) device whose a depletion region is controlled according to an inputted electrical signal to perform integration and leakage.

A Bifidobacterium animalis subsp. lactis for reducing Abeta42 deposition and an application, a name of the Bifidobacterium animalis subsp. lactis is Bifidobacterium animalis subsp. lactis IOB-LO7, and classified as Bifidobacterium animalis subsp. lactis, the Bifidobacterium animalis subsp. lactis IOB-LO7 is preserved in the General Microbiology Center of the China General Microbiological Culture Collection Center (CGMCC) on Dec. 23, 2021, with the collection number of CGMCC No. 24185. By reducing the levels of Aβ42 in the cerebral cortex and hippocampus, clearing Aβ amyloid plaques, improving communication between neurons, reducing brain neuroinflammation, and protecting nerve cells, it helps to restore cognitive function and improve Alzheimer's disease. Additionally, it can also improve gut microbiota imbalance caused by Alzheimer's disease.

This invention belongs to the field of biomedicine and relates to a biomarker for the early diagnosis of Alzheimer's disease and its application. Specifically, it discloses a biomarker for the early diagnosis of Alzheimer's disease, namely the Maf1 gene or its protein. High expression of the Maf1 gene or its protein in neuronal cells indicates that the test subject is in a high-risk group for Alzheimer's disease. This invention provides a new diagnostic and therapeutic target for AD.

A RELU neuron chip circuit belongs to the field of chip circuits, and is characterized by comprising a vector summation circuit, a shift circuit, a subtraction circuit, a logic judgment circuit, an input port and an output port, the input ports comprise a data input port, a weight input port, a bias data input port and a data counting port; the output port comprises a data output port and a logic output port; by directly realizing the function of the neurons on the chip circuit, when a neural network system calls a certain neuron to carry out corresponding function calculation, data input and output can be directly carried out at the bottommost circuit level, so that a large amount of data cross-layer conversion time is saved. Only after the whole neural network completes one time of complete training or judgment operation, the operation result of the bottom layer circuit can be transmitted to the high layer from the bottom layer circuit in a cross-layer mode and displayed in a human eye recognizable mode. Therefore, the overall operation performance of the neural network system is improved.

The invention provides a synaptic connection processing method and system for a brain-like computing network, and the method comprises the steps: decoding a pulse signal through a pulse address decoding module, and obtaining a head address and data length information of a corresponding target synaptic index; reading a corresponding target synaptic index from a memory through a synaptic index DMA module according to the initial address and data length information of the target synaptic index; decoding the target synaptic index through a pulse address decoding module to obtain a first address and a number of corresponding target synaptic connections; reading all target synaptic connections from a memory through a synaptic index DMA module according to the initial addresses and the number of the target synaptic connections; and obtaining target neurons corresponding to the target synaptic connections and corresponding synaptic connection weights through a synaptic connection distribution module, and sending the synaptic connection weights to the corresponding target neurons in the neuron processing system. According to the method, the indexing and distribution of the synaptic connections can be efficiently and reliably realized.

The present application relates to a kind of neural network global one-time structured pruning method, system, equipment and medium, wherein, method includes: by single forward propagation, in the neural network to be pruned using calibration dataset and parallelly capturing the input activation tensor of all target layers;According to input activation tensor, the intermediate neuron weight of each target layer is synchronously calculated differential entropy index and amplitude response intensity, and after normalization and fusion, static global importance atlas is formed;According to the importance threshold value determined according to preset pruning rate, based on global importance atlas, one-time generates the index set of all global pruning to be pruned whose mixed importance score is below importance threshold value;Based on index set, the weight matrix of each target layer is executed one-time physical structured pruning.By doing so, the present application generates static global importance atlas by single forward propagation and parallelly capturing activation tensor, and completes no-mask one-time pruning by physical structured pruning.

The invention belongs to the technical field of biological medicine, and particularly relates to an extracellular vesicle drug delivery system, a preparation method thereof and application of the extracellular vesicle drug delivery system in treatment of Alzheimer's disease. According to the invention, one or more of stilbene glucoside, emodin, kaempferol, resveratrol and quercetin are used for regulating and controlling the expression of miRNA (micro Ribonucleic Acid) related to the disease progress in the extracellular vesicles, and the result shows that one or more of miR-let-7c-5p, miR-486-5p, miR-132-3p, miR-160b and miR-129-5p is remarkably up-regulated, and one or more of miR-342-3p, miR-16-5p and miR-125b-5p is remarkably down-regulated. The regulated extracellular vesicles can be used for treating Alzheimer's disease, relieving Tau protein abnormal phosphorylation, reducing amyloid protein beta deposition, inhibiting microglial cell activation, protecting neuronal cells and relieving memory deficits.

The invention relates to the technical field of working face mine pressure multi-parameter intelligent monitoring and early warning systems, and particularly discloses a working face mine pressure multi-parameter intelligent monitoring and early warning system. The system comprises a multi-source sensing data acquisition device, a pulse sequence coding unit, a bionic pulse neural network early warning model, a dynamic attention regulation and control module and an early warning information output terminal, multi-source heterogeneous data such as hydraulic support resistance, micro-seismic energy and acoustic emission frequency are converted into a pulse event sequence similar to a biological neural signal, a rock stratum instability precursor is identified by using a bionic model simulating a neuron discharge mechanism and synaptic plasticity, key signals are self-adaptively strengthened in combination with a dynamic attention mechanism, and noise is suppressed. According to the technical scheme, multi-parameter coupling characteristics can be captured with high sensitivity, the early warning accuracy and advance are improved, the false alarm rate is reduced, and reliable guarantee is provided for deep coal mine safety.