66 results about "Neuronal activation" patented technology

The invention discloses a federated learning backdoor defense method based on pruning and fine tuning in the technical field of artificial intelligence and network security, the method realizes defense through two core mechanisms of dynamic pruning and gradient constraint fine tuning, and the method comprises the following steps: firstly, calculating a sensitivity score based on a neuron activation frequency and a weight outlier degree; dynamically identifying and cutting redundant neurons utilized by a backdoor, and blocking an abnormal activation path; secondly, gradient direction consistency detection and amplitude constraint are introduced in the fine tuning stage, and a malicious client is inhibited from reconstructing a back door through an abnormal gradient; the server continuously purifies model parameters and enhances robustness by cyclically executing pruning, fine tuning and aggregation operations; the method does not need to depend on an extra clean data set, strictly follows a federated learning privacy protection principle, reduces communication overhead through lightweight pruning, maintains main task performance in combination with gradient constraint, is suitable for a federated learning scene in which edge equipment participates, and effectively balances a defense effect and model stability.

The invention relates to the technical field of artificial intelligence model optimization, and discloses a multi-granularity dynamic pruning method and system for a generative AI model. The system comprises a model state acquisition module which acquires output of a middle layer in real time through a probe, and constructs a feature information set of neuron activation distribution, weight matrix norm and connection topology; the sparseness evaluation module outputs a sparseness risk value based on the feature information set, the initial sparseness parameter and the real-time computing resource state; a pruning planning collaborative analysis screening module generates a multi-granularity pruning scheme when the risk value exceeds a threshold value, calculates a collaborative interference value of precision recovery operation and screens an optimal scheme; and the pruning strategy execution feedback module updates model parameters, generates logs and feeds the logs back to the management terminal. The system realizes adaptive structure optimization and resource scheduling of the generative model in the reasoning process.

The invention relates to the technical field of geographic information systems, and discloses a geographic information data processing method and system.The geographic information data processing method comprises the steps that geographic data are packaged into self-activation data nodes; monitoring a network state in real time through a network environment perception function; constructing an activation threshold function based on a neuron activation principle to determine data transmission; carrying out multi-scale analysis on the data stream by applying wavelet transform to realize self-adaptive compression; predicting hotspot distribution dynamic scheduling resources through space-time analysis; an edge-fog-cloud three-level cooperative processing architecture is constructed to realize distributed processing; according to the method, the technical problem of geographic information data transmission and processing in an unstable network environment is solved, the overall performance and the adaptive capacity of the system are improved through a multi-level adaptive mechanism, and the method has wide application value.

The invention provides a visual motion function test system and method based on full-view surround synchronous visual stimulation, and is applied to the technical field of medical data processing. Visual stimulation parameters and rotating bar motion mode parameters are combined, the rolling direction of a bar and the rotating direction of a rotating bar are synchronously matched, a synchronous presentation scene of visual stimulation and motion tasks is established based on a four-screen linkage control technology and a rotating bar-screen adaptive design, and full-view stimulation constraint conditions are generated; processing is carried out based on visual stimulation parameters, rotating rod reference data, synchronous control logic and dynamic parameter combination constraint conditions, and rod time, rotating speed during falling and motion trail data are collected; processing the collected kinematics data, and combining functional parameters including visual state grouping and brain region neuron activation counting to generate motion balance ability and brain function associated data; and processing by combining normality test and an inter-group statistical method to generate a vision-motion function evaluation result.

A transcranial magnetic stimulation (TMS) treatment system is provided. The system includes a sensor device that senses EEG signals from a subject through one or more leads and a server device configured to receive EEG data corresponding to the subject. The server includes an analysis module configured to process the EEG data and determine a personalized resonant brain frequency and a minimum neuronal activation threshold of the subject based at least in part on EEG data corresponding to one or more leads of the sensor device. The analysis module is also configured to determine a TMS treatment protocol where the treatment protocol includes at least a frequency based on the personalized resonant brain frequency and an amplitude based on the minimum neuronal activation threshold. The system also includes a treatment device configured to deliver a TMS treatment to the subject based on the TMS treatment protocol received from the server.

The invention discloses a method and a device for decoupling a slow-thinking area and a quick-response area in a large language model. The method comprises the following steps: constructing a slow thinking prompt and a fast response prompt, respectively inputting the slow thinking prompt and the fast response prompt into a large language model, and recording a neuron activation value of each linear layer of the large language model; calculating an activation value difference vector based on the neuron activation values corresponding to the slow thinking prompt and the fast response prompt, and obtaining a prompt difference vector based on the activation value difference vector; iteratively processing a plurality of reasoning problems, obtaining positive value position intersection of prompt difference vectors in all reasoning problems, and identifying a key neuron set related to slow thinking; and constructing a gradient mask matrix based on the position information of the key neurons in the key neuron set, and performing fine tuning on the key neurons by adopting a mask mode based on the gradient mask matrix to obtain an optimized large language model. The technical problem that an existing model is poor in reasoning ability is solved.

The invention discloses a hardware acceleration system and method supporting a pulse-gated recurrent neural network, and relates to the technical field of neural network hardware acceleration, and the system comprises a PC terminal and an accelerator. The accelerator comprises a UART (Universal Asynchronous Receiver / Transmitter) module, a data composer, a storage module, a controller module, a calculation module and a gating calculation core, the PC terminal is connected with the UART module; the gating calculation core is connected with the data composer; and the controller module and the storage module are respectively connected with the UART module, the data composer, the calculation module and the gating calculation core. Based on the above scheme, different coding modes can be selected at a PC terminal to adapt to different data types, and a pulse neuron activation function and a logic gate are utilized to perform calculation heterogeneous design in a calculation module and a gating calculation core, so that a high-performance configurable hardware acceleration calculation architecture for the pulse gating recurrent neural network is realized.

The invention relates to a technique for improving confidence estimates associated with neural networks. The technique involves computing neuron activation statistics during training, evaluating neuron activations during inferencing and determining how the activations compare with the previously computed statistics (e.g. whether prediction activations are within the bounds of the training activation statistics). The comparison may be used to compute a confidence value for the neural network.

There are provided systems and methods for data privacy protection and removal for artificial intelligence model training and deployment. An online transaction processor or other service provider may provide computing services and platforms to entities, which may include use of machine learning (ML) models including large language models (LLMs). To comply with data privacy protections and copyright enforcement, a system may provide unlearning of content from ML models. The system may receive a request to unlearn a content and, after verifying the request is valid, identify the content used for during training of or inferencing by an ML model. The system may then map the content to concepts and correlate those concepts with ML model outputs using projections in a vector space. Based on the mapped concepts and outputs, neuron activation of the ML model may be analyzed to identify a negation vector and perform selective parameter dampening.

The invention relates to the technical field of capacity collaboration, and discloses a capacity collaborative optimization method and system based on a knowledge graph, and the method comprises the steps: constructing a capacity collaborative decision model based on the knowledge graph; designing a distributed homomorphic neural computing framework; developing a pulse neural network encoder; realizing a pulse neural network inference mechanism in a homomorphic encryption domain; designing a neuron activation mechanism based on a threshold value; constructing an inter-enterprise security communication protocol; developing a calculation result verification mechanism based on zero knowledge proof; a result decryption mechanism for realizing separation of rights and responsibilities; designing a special neuromorphic hardware acceleration chip; constructing a feedback updating mechanism of the decision result and the knowledge graph; according to the method, an enterprise can participate in collaborative decision-making on the premise of not leaking sensitive data through a secure computing technology, and meanwhile, the energy consumption and delay of complex computation are reduced by utilizing a neuromorphic computing architecture.

The invention discloses a road target detection method and device based on multi-sensor fusion, and belongs to the technical field of target detection, and the method comprises the steps: collecting the multi-modal sensor data of a target road and the position information of each sensor; based on the position information of each sensor and the multi-modal sensor data, analyzing a space-data communication relationship among the sensors, and constructing a sensing adjacency matrix; extracting spatio-temporal characteristics of the multi-modal sensor data based on the sensing adjacency matrix, generating a spatio-temporal characteristic set, and obtaining a neuron activation value of each calculation path in real time; performing redundancy analysis on each calculation path according to the neuron activation value of each calculation path, and screening out a reserved path; performing redundancy removal processing on the space-time feature set based on each reserved path to obtain a sparse space-time feature set; and performing data fusion on the sparse space-time feature set to obtain a target detection result of the target road. By implementing the method and the device, the problem of low target detection accuracy in the prior art can be solved.

The invention discloses an optimal path planning method of a pulse coupling neural network based on double constraints. The method comprises the following steps: mapping a path planning environment to a DC-PCNN network; a DC-PCNN neural network model is constructed, and all neurons are initialized; activating a target neuron, and recording the current neuron as a father node; calculating an exponential decay function of the torque deviation, and multiplying the calculated value as a penalty factor by an update item of the internal activity item; calculating a gravitational function value of the flow field constraint, and using the calculated value to update a current neuron dynamic threshold value; comparing the internal activity item with a dynamic threshold; when gt; if yes, activating the neuron, and recording the current neuron as a father node; repeating the steps S4-S6 until the initial neuron is activated; and backtracking all activated nodes, and planning an optimal path. According to the method, the search efficiency is remarkably improved while the path optimality is ensured.

The application discloses a heterogeneous federated training and reasoning method resisting backdoor attacks, and comprises cloud services. The method embeds a BN layer to record the variance and mean of neuron activation data in training when a global model is generated, then selects a benign client by measuring the neuron activation distribution through KL divergence and Bottom-K voting in the training stage, finally inputs a task data set into the trained global model in the reasoning stage, so as to obtain the probability distribution of each category, and then aggregates all probability distributions by using the maximum value or the mean value to obtain a reasoning result. The application solves the problems that the existing defense methods have poor defense effect against backdoor attacks under data heterogeneity, most backdoor attack detection methods need an additional data set, and heterogeneous federated learning does not study the inhibition of backdoor attacks in the reasoning stage.

The application discloses a video recommendation historical bias mitigation method based on bias attention, which comprises the following steps: (1) data preprocessing; (2) searching for bias neurons; (3) calculating bias masks; (4) calculating bias attention masks; (5) outputting bias attention masks; (6) repeating steps (2), (3), (4) and (5) for each layer until all layers are processed by the bias operation and the output layer is ended. The application avoids the problem of high calculation complexity caused by massive data; the bias attention suppresses the activation output of the bias neurons without changing the normal neuron activation value, and can effectively offset the bias behavior caused by the bias neurons.

The lithium ion battery SOC prediction method based on the CEEMDAN and the optimized BiLSTM comprises the following steps: decomposing preprocessed operation data by adopting adaptive noise complete ensemble empirical mode decomposition CEEMDAN to obtain a plurality of intrinsic mode functions (IMF) and a residual component; constructing a BiLSTM model, wherein the model comprises a forward LSTM unit, a backward LSTM unit and an output layer; a dropout optimization algorithm is introduced into the BiLSTM model, and a combination of a two-parameter sigmoid activation function and a softsign activation function is adopted as a neuron activation function; and inputting the intrinsic mode function IMF and the residual component into the optimized BiLSTM model, optimizing model parameters through training, and predicting the state of charge SOC of the lithium ion battery by using the trained model. Aiming at the problems of weak data nonlinear processing capability, model overfitting, gradient dispersion and the like in the existing SOC estimation method, the invention provides a lithium ion battery SOC prediction method based on CEEMDAN and BiLSTM optimization, so as to improve the SOC prediction precision and model stability.

The invention relates to the technical field of deep learning, in particular to a neural network model optimization method and device, a terminal and a storage medium. Carrying out robustness test on the neural network model through disturbance samples in the disturbance sample set, and judging whether the disturbance samples are target disturbance samples or not according to a test result; if yes, activation degree distribution information of the target disturbance sample and the target standard sample is obtained, and the activation degree distribution information is used for reflecting the activation degree of each neuron when the neural network model carries out reverse transmission; and optimizing the neural network model according to the activation degree distribution information of the target disturbance sample and the target standard sample. The activation condition of each neuron is obtained in a layer-by-layer reverse transmission mode, so that the neural network model has interpretability. Model optimization is carried out by analyzing the difference of the neuron activation conditions of the neural network model under the target standard sample and the target disturbance sample, and the robustness of the neural network model can be effectively improved.

The application discloses a federal fair framework training method and device based on neuron activation difference. In the process of federal learning, in addition to requiring the client to upload the local model update parameter, the server also requires the client to upload the activation frequency of neurons, and the server evaluates the contribution degree of each client by calculating the sparsity of the neuron activation frequency of the client, so as to issue different quality global models to various clients.

The present invention relates to the field of artificial intelligence explainable technology and discloses a method for preventing cue word attacks in a large language model based on neuron correction, comprising: constructing an attack cue word dataset, rewriting the attack cue words to obtain a control cue word dataset composed of control cue word data; constructing probes to obtain the output values ​​of all neurons in the large language model during inference, thereby obtaining the neuron activation state of the large language model; inputting the attack cue words and corresponding control cue words into the large language model, summarizing the obtained neuron activation state table, and obtaining additional activated neurons corresponding to the attack cue words; inputting the cue word attack data into the large language model, and forcibly setting the output of one or more of the additional activated neurons to zero. By correcting key neurons, the model performance and prevention effect are balanced, and the security of the large language model is improved.

The application provides a visual motion function test system and method based on full-view surround synchronous visual stimulation, and applies to the technical field of medical data processing. The visual stimulation parameters and the rotating rod motion mode parameters are combined, the strip grid rolling direction and the rotating rod rotating direction are synchronously matched, the synchronous presentation scene of the visual stimulation and the motion task is established based on the four-screen linkage control technology and the rotating rod-screen adaptive design, and the full-view stimulation constraint condition is generated; the visual stimulation parameters, the rotating rod reference data, the synchronous control logic and the dynamic parameter combination constraint condition are processed, the rotating rod time, the falling rotating speed and the motion trajectory data are collected; the collected kinematic data are processed, the function parameters including the visual state grouping and the brain region neuron activation count are combined, the motion balance ability and brain function correlation data are generated; and the visual-motor function evaluation result is generated according to the combination of the normality test and the inter-group statistical method.

The present invention discloses a CRH PVN The invention relates to a method for constructing a mouse model of liver depression-type breast cancer with neuronal activation and its application, and belongs to the technical field of animal model construction. The method comprises the following steps: (1) injecting 140-160 nl of liquid containing both rAAV-CRH-CRE-WPRE-hGH polyA and chemical genetic activation virus rAAV-hM3D(Gq)-mCherry into the PVN brain region of CRH-IRES-Cre mice, wherein the titers of the two viruses are 2.4-2.6×10 12 pfu; (2) 25-30 days after virus injection, all mice were implanted with breast cancer cells, and CRH was obtained 38-42 days after breast cancer cell inoculation. PVN The method for constructing a mouse model of liver depression type breast cancer with neuron activation is simple to operate and has good repeatability, and can overcome the disadvantage of large individual differences in animals.

The invention belongs to the technical field of taVNS parameter adjustment, and relates to a taVNS closed-loop regulation and control method for high-inflammation ARDS, and the method comprises the steps: obtaining a neuron activation index through employing a dual-frequency auricular conchae impedance analysis technology, inputting the time series data of the neuron activation index into an inflammation risk prediction model, obtaining an inflammation risk score, and achieving the quantitative evaluation of an inflammation state; in combination with a comparison result based on an inflammation risk score and a set threshold value, a corresponding frequency control parameter is dynamically output in a multi-decision mode, and the current intensity is adjusted in combination with a real-time neuron activation index, so that a stimulation parameter can be adjusted in real time along with the change of an inflammation state; after stimulation is applied, by monitoring the neuron activation index change rate and the heart rate variability high-power index, the stimulation frequency is corrected within the preset time, a closed loop of state perception, parameter adjustment and effect feedback to parameter correction is formed, it is ensured that the stimulation parameters are accurately matched with the inflammation state all the time, and the stimulation effect is improved. Therefore, the target of dynamically adjusting parameters according to the inflammation state is achieved.

Disclosed herein are inhibitors of aminocarboxymuconate semialdehyde decarboxylase (ACMSD), pharmaceutical compositions and methods for using these compounds to treat or prevent one or more ACMSD-related diseases such as, for example, muscle structure disorders, neuronal activation disorders, muscle fatigue disorders, muscle mass disorders, beta oxidation diseases, metabolic diseases, cancer, vascular diseases, ocular vascular diseases, muscular eye diseases, liver diseases, inflammatory diseases or renal diseases.

The application discloses a multi-style text migration method based on a shared neuron activation modulation network, and belongs to the technical field of natural language processing, large language model neuron regulation and text style migration. The method first constructs a style dimension and neuron mapping relationship, and divides exclusive neurons and shared neurons; directional activation modulation is performed on the exclusive neurons, and a embedded shared neuron activation modulation network (SNAR) is constructed; through style interaction embedding, joint coding, context modeling and content cross attention decoding, the shared neuron activation distribution is dynamically reorganized, and multi-style representation conflicts are eliminated; finally, the activation values after modulation of the two types of neurons are fused, and multi-style text is generated through model reasoning. The application can efficiently realize cross-dimension multi-style collaborative migration without fine-tuning the large language model, solve the style conflict, semantic deviation and activation competition problems of existing methods, and is suitable for human-computer interaction, content creation and other scenes.

The invention discloses a method for measuring defective neurons of a convolutional neural network model of a migrated brain neural network, and belongs to the technical field of intelligent measurement and testing. According to the method, on the basis of a cognitive activation hypothesis theory, accurate measurement of internal defects of a model is realized by constructing a'traceability chain 'reflecting a defect propagation path and positioning'suspicious neurons' based on differential statistics; the method specifically comprises the following steps: detecting abnormal neurons, and generating and combining high-weight paths to construct a tracing chain; meanwhile, key suspicious neurons are counted and positioned by comparing neuron activation differences under similar input. According to the method, weak links in a model decision-making mechanism can be accurately revealed, the efficiency and interpretability of defect positioning are remarkably improved, and an effective technical support is provided for enhancing the robustness of the model in a safety-critical scene.

Disclosed herein are aminocarboxymuconate-semialdehyde decarboxylase (ACMSD) inhibitors, pharmaceutical compositions, and methods of using these compounds to treat or prevent one or more ACMSD-related diseases, such as a muscle structural disorder, a neuronal activation disorder, a muscle fatigue disorder, a muscle mass abnormality disorder, a beta-oxidation disease, a metabolic disease, a cancer, a vascular disease, an ocular vascular disease, an ocular muscle disease, a liver disease, an inflammatory disease, or a kidney disease.

The invention discloses a continuous learning method based on bidirectional dynamic sparseness, and relates to the technical field of machine learning, and the method comprises the steps: S1, carrying out the initialization of neurons and connections in a neural network; s2, multi-dimensional neuron activity assessment: performing neuron activity assessment from three dimensions of neuron activation values, gradients and importance of front and back connections of neurons, thereby improving accuracy and flexibility of neuron activity assessment in the model; s3, bidirectional dynamic sparse training: initializing a mask function according to the weight of the neuron and the activity evaluation value; carrying out sparse optimization on the network topology structure by utilizing the mask function; the mask function and the neuron weight are updated according to the mask loss and the neuron weight loss of the mask target function, and it is ensured that the optimal mask sparse strategy can be obtained in a complex task scene. And the plasticity of the model in the continuous learning process is improved.

A method for generating a device-specific artificial intelligence model includes obtaining, by a hardware device, unique information of a specific device, generating, by the hardware device, a neuron activation matrix based on the unique information, setting, by the hardware device, a degree of activation for neurons by applying the neuron activation matrix to neurons of a neural network model, and performing, by the hardware device, fine-tuning on the neurons for which the degree of activation has been set using a positive dataset.

The invention belongs to the field of artificial intelligence, discloses a deep neural network robustness enhancement method for fragile neuron adversarial training, and aims to solve the problem that the natural precision is severely reduced in an existing adversarial training method. According to the method, the concept of a positive value pushing sample is introduced, and generation of the positive value pushing sample with certain aggressiveness is guided by accurately positioning and activating fragile neurons of the model. A target optimization layer is identified by using neuron activation difference indexes of a clean sample and an adversarial sample, and a neuron activation optimization function is designed. And iteratively generating a positive value pushing sample through a gradient optimization method, so that the positive value pushing sample explores an area which is not fully covered in the DNN in a feature space. And the difference between the generated sample and the clean sample in the data manifold is smaller than that of the traditional adversarial sample. Experiments prove that the method keeps the natural precision of the model to the greatest extent while improving the robustness of the model, and is suitable for practical application with high requirements on the natural precision.

The invention discloses a coarse-to-fine land utilization change intelligent detection method and system. The method comprises the following steps: making a land utilization scene classification data set; according to the classification precision, carrying out adaptability distinguishing on the CNN model to complete screening and carrying out model fine tuning; determining a land utilization scene category through a multi-CNN collaborative scene category identification mechanism; spectral change intensity information is extracted based on an RCVA method; texture change intensity information is extracted based on a GLCM method; combining a scene classification result to complete the selection of a training sample; constructing a DBN model; defining joint probability distribution of an explicit layer and a hidden layer of the RBM based on an energy function; determining a neuron activation probability based on the structural characteristics of the RBM; fitting training data through a maximized log-likelihood function; and completing model training. And the processing efficiency, the monitoring precision and the comprehensive application value of change information are effectively improved.

Embodiments of the present application provide a neuron activation prediction method and an electronic device. The neuron activation prediction method comprises: performing local calculation based on input features of a current layer of a neural network to generate a local vector for neuron activation prediction; performing activation prediction and screening operation on a plurality of candidate neurons in the current layer of the neural network based on the local vector to obtain a target neuron set and a sparse execution index table corresponding to the target neuron set; performing multiplication and addition calculation and activation function processing on target neurons in the target neuron set based on the sparse execution index table to obtain a sparse activation output result of the current layer of the neural network; determining prediction error information based on the sparse activation output result and an actual activation result, and determining a neuron screening strategy for a next layer of the neural network based on the prediction error information, which can reduce the memory bandwidth demand and energy consumption overhead in the inference process of the neural network.