1072 results about "Model architecture" patented technology

A method includes obtaining an image, the image associated with a mask corresponding to a portion of the image, generating a plurality of images based on the image and the mask, each image of the plurality of images depicting a different color in the portion of the image corresponding to the mask, executing a machine learning model to generate an image performance score for each of the plurality of images, ranking the plurality of images according to the image performance scores for the plurality of images, and generating a record comprising one or more images of the plurality of images based on the rankings of the plurality of images.

The invention relates to the technical field of power systems and automation thereof, in particular to a distribution network auxiliary decision-making method and system considering source load fluctuation relevance and a medium. The method comprises the following steps: firstly, collecting related information of a distribution network area, quantifying a synchronization and hysteresis association rule of multi-source heterogeneous data fluctuation, and constructing a composite feature vector and a standardized risk perception data set; defining a state space and an action space of a reinforcement learning algorithm based on the composite feature vector, and realizing auxiliary decision-making optimization of the distribution network; constructing a scene feature library, calculating the fluctuation relevance similarity between a new scene and a historical scene, and multiplexing a deep reinforcement learning model architecture and carrying out transfer learning; building a power grid digital twinborn simulation platform, designing evaluation indexes, generating candidate schemes, deducing the candidate schemes, selecting recommendation strategies and storing the recommendation strategies in a strategy knowledge base.

The invention discloses an AI-based automatic financial information processing method, and relates to the field of financial automation, and the method comprises the steps: achieving the automatic collection and storage of structured and unstructured data through the access of enterprise multi-source financial data; systematic preprocessing is carried out on the collected multi-source heterogeneous financial data, and a unified and high-quality financial data set is constructed; based on natural language processing and a knowledge graph technology, performing text semantic understanding, transaction automatic classification, field standardization and label generation on the cleaned and integrated financial data; comprehensively quantifying enterprise operation and financial performance based on the structured transaction data and the semantic annotation result; based on historical financial indexes, establishing a multi-model architecture to predict key financial variables; and based on the structured data, the prediction result and the historical rule, identifying potential financial abnormity and risk behaviors, and realizing intelligent early warning. According to the method, the intelligence, the real-time performance and the accuracy of financial information processing can be remarkably improved.

The invention discloses a steel surface defect detection method and device based on YOLO11n improvement, and relates to the technical field of defect detection.The method comprises the steps that a to-be-detected steel surface image collected by an image collection device is obtained; inputting the to-be-detected steel surface image into a defect detection model obtained by training the improved YOLO11n model architecture to obtain surface defect detection information of the to-be-detected steel surface image; wherein a backbone network of the improved YOLO11n model architecture comprises a C3k2NAM module formed by a C3k2 module integrated with an NAM dynamic attention mechanism, and an Neck network introduces an NCB convolution block, an MSDA attention mechanism and a CoT Attention module. The problem that accurate identification of a multi-scale target cannot be realized in steel surface defect detection in the prior art is solved.

The invention discloses a privatized AI question and answer method, system and device and a medium, and relates to the technical field of artificial intelligence, and the method comprises the steps: collecting a file archive data set, and constructing a local digital knowledge base; selecting an AI semantic large model architecture; performing annotation enhancement on a local digital knowledge base by using an AI semantic large model architecture, and constructing a knowledge base AI question and answer model; performing permission marking on the local digital knowledge base based on a user access control rule, and establishing a knowledge base retrieval mechanism; and integrating and fusing the knowledge base retrieval mechanism and the knowledge base AI question-answering model to generate a retrieval enhanced AI question-answering model, performing semantic question-answering retrieval on the request question information, and outputting a user request question-answering result. The technical problem that in the prior art, data security, intelligent question answering and information retrieval efficiency are insufficient is solved, and the technical effect of improving intelligence and data access control of the question answering system is achieved.

The invention discloses a typhoon wave forecasting method based on integrated machine learning. The method comprises the following steps: firstly, integrating historical typhoon wave data, meteorological data and marine environment data; preprocessing the data, including integration, cleaning, vacancy filling and standardization, and performing multi-source data completion by adopting a K-nearest neighbor algorithm and a spline interpolation method; secondly, screening key characteristic parameters through a Pearson's correlation coefficient, and reinforcing nonlinear correlation representation in combination with a mutual information method; then, constructing an integrated prediction model containing an LSTM (Long Short Term Memory), an XGBoost (X Goose Boost) and a Transform; and finally, dividing a training set and a verification set by adopting a dynamic time sequence division strategy, optimizing model hyper-parameters, and completing training and testing of the typhoon wave height prediction model. According to the method, the data sparsity problem is solved through multi-source data fusion and feature selection optimization, the generalization ability is improved through an integrated model architecture, and compared with a traditional single model, the training period is remarkably shortened, and the forecasting precision and timeliness are improved.

The invention provides an automatic driving vehicle multi-mode track prediction method based on Transform and end point induction. The method comprises the following steps: firstly, extracting historical track characteristics of surrounding vehicles, historical track characteristics of a target vehicle, relative position information and scene map characteristics; carrying out local and global fusion on the extracted features by utilizing Transform, and efficiently and comprehensively fusing the features of each element in the scene; secondly, predicting a multi-modal trajectory end point through a dynamic weight multi-layer perceptron (MLP), simplifying a model architecture and reducing the quantity of model parameters; and finally, performing trajectory refinement by using attention mechanism interaction end point information, and outputting multi-modal trajectory prediction, so that the predicted future trajectory of the vehicle is reasonable and accurate. The method not only can improve the multi-modal trajectory prediction accuracy of the autonomous vehicle, but also can improve the model prediction efficiency.

The invention relates to the technical field of digital information transmission, and provides a communication network security situation prediction method and system based on big data, which break through the limitation of single data in the aspect of data fusion, integrate four large classes and 12 subclasses of multi-source heterogeneous data, and combine a dynamic weighting mechanism to make feature extraction more comprehensive and accurate; on the aspect of model architecture, CNN-Bi-LSTM-Attention three-level fusion and PSO optimization are adopted, spatial and temporal features are effectively captured, the convergence speed is increased, the prediction accuracy is high, and the false alarm rate is low; in the risk assessment aspect, a layered assessment system is constructed, and early warning is realized in combination with a dynamic threshold value and Monte Carlo simulation; the model training module innovatively adopts a federated learning mode, and the model generalization ability is improved while data privacy is guaranteed; the prediction analysis module deploys an optimization hybrid model, supports high-concurrency prediction and is short in response time; the visual decision-making module provides three-dimensional visualization and geographical drilling functions, and output data can be seamlessly connected with a third-party platform.

The invention discloses a slope early warning method and system based on deep learning, and particularly relates to the technical field of slope early warning, and the method comprises the steps: S1, multi-source data collection, S2, dynamic graph construction, S3, meta-learning model initialization, S4, space-time fusion prediction, S5, dynamic risk assessment, and S6, graded early warning triggering. Through multi-modal data fusion, an innovative model architecture and an intelligent early-warning mechanism, the slope early-warning capability can be remarkably improved, multi-source data are fused, a cross-modal attention mechanism is utilized, the slope state is comprehensively and accurately reflected, the early-warning accuracy is improved, a dynamic graph structure is constructed to be combined with a meta-learning engine, different slopes are adapted, continuous optimization can be achieved, and the early-warning capability of the slope is improved. Meanwhile, a scientific grading early warning system is established, a historical case library and related equipment are linked, resources are efficiently allocated, life and property safety is guaranteed, and disaster losses are reduced.

Systems and methods disclosed herein generate validated responses using artificial intelligence (AI)-based models. The system obtains / receives an output generation request (e.g., from a graphical user interface (GUI)) that can include a document set and a query set. The system classifies the query set by partitioning it into multiple query subsets and assigning a complexity score. Based on the classification, the system generates a computational workflow set using a first AI model set to retrieve a resource set responsive to the query set. The system executes the workflow using a second AI model set (the same as or different from the first AI model set) and validates the retrieved resources against predefined criteria (e.g., rules, guidelines). If the resources satisfy the criteria, the system generates a response using a third AI model set. The system can display a graphical layout on the GUI showing the request, retrieved resources, and / or generated response.

A machine-learned architecture may predict a set of spatially-diverse paths that an object may take in the future. The paths generated by this architecture may be time-invariant (e.g., not identifying a time at which the object may occupy a position along one of these paths) but can be used by a second machine-learned model to predict progress in time along these paths. This segregation of the spatial paths and progress in time along the paths improves the accuracy of the ultimate prediction and better captures rare object behavior.

The invention relates to the technical field of artificial intelligence, and discloses a financial report analysis method, and the method comprises the steps: carrying out the data preprocessing of historical financial association data and historical business operation data, and obtaining to-be-analyzed historical data; performing multi-order feature engineering processing on the to-be-analyzed historical data to obtain historical feature vector data; constructing an initial financial analysis large model, and training the initial financial analysis large model by using the historical feature vector data to obtain a trained financial analysis large model; accessing the trained large financial analysis model into a target system, and triggering the large financial analysis model to operate; and driving the large financial analysis model to perform multi-dimensional semantic analysis and quantitative reasoning on the to-be-analyzed financial report data to obtain visual financial report analysis result data. The method can be applied to internal financial statements of enterprises with businesses of science and technology finance, medical health, old-age care and the like, and the financial statement analysis efficiency and comprehensiveness can be improved through the multi-modal data fusion and knowledge enhancement large model architecture technology.

The invention relates to the technical field of fault diagnosis, in particular to a wind driven generator fault diagnosis method and system based on Mamba-ResNet. The method comprises the following steps: carrying out feature extraction and feature fusion by utilizing preprocessed data, namely constructing adaptive window short-time Fourier transform (AW-STFT) to carry out dynamic time-frequency resolution analysis, carrying out parallel feature extraction and constructing a multi-dimensional heterogeneous feature vector, and carrying out a cross-modal adaptive gating fusion mechanism based on a bidirectional cross gating unit; the method comprises the following steps: constructing a Mamba-ResNet hybrid deep network model architecture; performing model training on the constructed network model architecture; and performing fault diagnosis on the wind driven generator by using the trained model architecture. A tedious manual feature design process in a traditional method is avoided, and the automation level and adaptability of a diagnosis system are remarkably improved.

The invention discloses a method for constructing a nursing field text annotation corpus based on deep learning, and relates to the field of artificial intelligence technology and medical information processing. Comprising a data collection and preprocessing step, a BERT-based nursing field entity recognition model construction and training step, an automatic labeling and post-processing step, a corpus construction and management step and an application and intelligent support step. According to the method, the problems of efficiency and accuracy of text labeling in the nursing field are solved, and the continuously developing and changing text data processing requirements in the nursing field can be better met. The entity recognition model architecture specially aiming at the text characteristics in the nursing field is constructed based on deep learning, and challenges can be effectively handled when complex and diversified nursing text data is processed. And meanwhile, by utilizing the automatic labeling capability of deep learning, the labeling efficiency is improved, the manpower and time cost is reduced, and a new direction, a new mode and new experience are provided for nursing informatization development.

The invention provides a ship trajectory prediction method based on a graph attention mechanism and an electronic chart, and relates to the technical field of intelligent ships. By fusing dynamic AIS trajectory data and static navigation channel geographic information, high-precision modeling and prediction of the future motion trajectory of the ship are realized. According to the method, a traditional time sequence-based trajectory modeling mode is expanded into multi-modal joint modeling, channel structure vector representation is introduced for the first time, structural information of a channel environment is extracted on the basis of a graph neural network, and deep fusion is performed on the structural information and historical trajectory data of a ship; and constructing a trajectory expression mode containing time, space and environment triplex semantics at the same time. In the aspect of model architecture, a graph attention mechanism is adopted to enhance the node representation capability in a channel sub-graph, and meanwhile, a time sequence modeling module is introduced to model a trajectory evolution rule, so that the prediction precision and generalization capability of the model in a complex water area environment are effectively improved, and high-quality modeling, interpretable prediction and intelligent support of the ship trajectory are realized.

The invention belongs to the field of hydrological data processing, and particularly discloses a runoff data interpolation model construction method and system. The method comprises the following steps: constructing a time sequence runoff data set on the basis of spatial and temporal distribution characteristics of hydrometric stations of a river basin; a space-time coupling interpolation model architecture is constructed and comprises an encoder and a decoder, the encoder comprises a Bi-LSTM module and a multi-head attention module, the Bi-LSTM module extracts forward and backward local time sequence characteristics of time sequence data layer by layer through a bidirectional information transmission mechanism, the multi-head attention module is used for capturing a space-time relationship in the time sequence data, and the time sequence data is subjected to time sequence data processing. The decoder adopts a mask self-attention mechanism to combine with a feature weight fusion module to dynamically correct a missing value; and performing model training and optimization on the interpolation model by using the time sequence runoff data set. According to the invention, interpolation of key time sequence dynamic information in the hydrological field can be realized.

Predicting the salience of one or more data entities to a particular (target) data entity from among a plurality of data entities may comprise generating a graph of the plurality of data entities and a machine-learned model architecture that predicts the salience of the one or more data entities output by the machine-learned model architecture using the graph. For example, the machine-learned model architecture may comprise a first machine-learned model for generating an embedding using the content of the target data entity, a second machine-learned model for generating a vector using the data type indicated by the target data entity, and a third machine-learned model (e.g., a graph neural network or other feed-forward neural network) for generating a contextual representation of the target data entity to which other contextual representations associated with the plurality of data entities may be compared (e.g., using Euclidean distance, cosine similarity, dot product).

The invention discloses a water and soil loss dynamic risk assessment system based on a remote sensing image, and relates to the field of water and soil loss risk assessment. The invention discloses a water and soil loss dynamic risk assessment system based on remote sensing images. The system comprises a data acquisition and preprocessing module, a dynamic feature extraction and analysis module, a hierarchical risk assessment model and a risk grading and modeling module, the method further comprises the following steps: acquiring a remote sensing image and historical soil erosion map data of a specific administrative region in a specific period through a data acquisition and preprocessing module; the dynamic feature extraction and analysis module is used for extracting land utilization types, vegetation coverage and quarterly change features of topographic factors through the acquired data; according to the method, a three-layer model architecture is arranged, static historical data, human activity disturbance and social exposure are decoupled and quantified, and the evaluation dimension is improved, so that the accuracy of an evaluation conclusion is improved.

The invention provides a motor defect identification method fusing time sequence space feature extraction and reinforcement learning. The method comprises the following steps: building a Transform-GAT model architecture T-GAT, and learning time sequence relevance and spatial topological structure features by the T-GAT to form a space-time composite representation vector; designing a dual-network architecture reinforcement learning weighted fusion mechanism of a strategy network and a value network, and after the strategy network and the value network of a parallel structure receive the composite vector, constructing a strategy function to select a defect type with the maximum probability; designing potential energy function quantization parameters, and constructing a reward function to generate a reward in combination with a difference value; taking a reward function as a target, training and optimizing model parameters of the dual-network architecture in an off-line manner, running a real-time decision in an on-line manner, and storing a tetrad to an experience pool to form a'perception-decision-feedback-update 'closed-loop mechanism; according to the method, motor defect identification is realized, downtime is reduced, and motor operation reliability and equipment operation efficiency are improved.

The invention relates to the technical field of computers, in particular to a server running state monitoring method and system and a medium. The method comprises the steps of obtaining time sequence operation and maintenance data of a system service; the method comprises the following steps: mapping an unstructured log text into a low-dimensional dense text vector by adopting an embedded learning method, splicing the text vector and a standardized numerical vector of a structured index to obtain a unified high-dimensional feature vector, and forming a cross-time vector database; constructing an AI model used for time sequence prediction, anomaly detection and cascade reasoning of classification decision based on a multi-model fusion architecture; training the AI model based on the vector database; and inputting the real-time feature vector into the trained AI model, and outputting to obtain a decision result of the system service operation state. Uniform expression of cross-modal features is effectively realized, cascade model architecture design is cooperated, the dynamic adjustment capability of the model and the accuracy of composite fault detection are improved, and rapid decision-making of fault types is realized.

The invention belongs to the technical field of computing resource scheduling, and particularly relates to an intelligent scheduling method and system for Huaan Atlas heterogeneous computing resources based on dynamic load awareness. The method comprises the steps that the real-time state of multi-dimensional hardware data is collected, and a basic data source is provided for subsequent steps; dynamically adapting tasks and hardware characteristics through a matching degree matrix, modeling aiming at various basic data, and constructing a state vector required by reinforcement learning; predicting a fault risk score through a lightweight prediction model deployed at each computing node; and a deep Q network is adopted as a model architecture, a state vector and a fault risk score are input, reinforcement learning training is performed through a reward function in a multi-target vector form, a final scheduling model is obtained, and a task allocation decision is output. The problems that in the prior art, the hardware state cannot be sensed in real time, hardware characteristic matching is ignored, consequently, the computing resource utilization rate is insufficient, and fault recovery is passive are solved.

The invention provides a medical image segmentation method based on an AFMHiFormer. The method comprises the steps that firstly, a multiple data enhancement module is provided, and the data distribution diversity is improved while the enhancement stability is guaranteed; secondly, a segmentation model AFHiMFormer is constructed, and the model architecture adopts a double-branch encoder and a multi-scale decoder; thirdly, a feature enhancement module is provided to construct a dynamic complementation mechanism of semantic enhancement and boundary modeling; fourthly, a multi-scale feature fusion module is introduced, multi-scale context information is captured through parallel hole convolution with different expansion rates, and self-adaptive fusion of global and local features is achieved; and fifth, a cross-scale fusion module is designed in the multi-scale decoder, so that the deep layer branch and the shallow layer branch are efficiently fused in a multi-level feature space. According to the method, the advantages of CNN and Transform are combined, dynamic fusion of local and global features is realized by providing a new module, and a remarkable performance advantage is shown in a medical image segmentation task.

The invention relates to a power equipment defect detection method and system. The method comprises the steps of firstly collecting historical defect image data of power equipment, and preprocessing to obtain a data set; s2, a deep learning model is built, the model architecture comprises a quantum neural network, a backbone network, a multi-scale feature fusion layer and a target position and category prediction layer, the deep learning model is trained by using the data set in S1, and a defect detection model is obtained; and finally, deploying the defect detection model to edge equipment, inputting the current target power equipment image into the defect detection model by the edge equipment, and outputting a power equipment defect detection result, thereby realizing power equipment defect detection. Compared with the prior art, the method has the advantages of being suitable for a complex inspection environment, accurate in detection, high in confidence degree and the like.

The invention provides a model training method and device, computer equipment and a computer readable storage medium. The method comprises the following steps: acquiring resource states corresponding to a plurality of hardware clusters included in a heterogeneous environment; determining a parallel strategy by taking load balancing as a target according to a model architecture of the hybrid expert model and a resource state; the parallel strategy comprises parallelism degree configuration and a target mapping relation between the hybrid expert model and each device in the hardware cluster; wherein the hybrid expert model comprises a plurality of Transform layers; each Transform layer comprises a plurality of expert models which are arranged in parallel; a plurality of expert models in the same Transform layer are mapped to different devices in the same hardware cluster; and deploying the hybrid expert model into a heterogeneous environment according to a parallel strategy, and performing distributed hybrid parallel training on the hybrid expert model in the heterogeneous environment.

The invention provides a code dynamic completion method for a large language model. The method comprises the following steps: constructing a mixed training data set; the method comprises the following steps of: initializing a Transform model architecture fused with probability space topological transformation; based on the mixed training data set, executing model training of collaborative optimization of the main target and the auxiliary target; in each step of decoding, the model calculates the probability distribution of the next token according to the current context; the generated probability distribution is restrained in an effective probability subspace through spherical probability mapping, and grammar error candidates are filtered in real time and the probability distribution is adjusted in combination with a tabu table mechanism; and dynamically selecting a sampling strategy according to the current decoding depth, sampling from the adjusted probability distribution to obtain a next code snippet, and finally generating a code completion suggestion conforming to abstract syntax tree rules and semantic constraints. According to the method, the probability distribution constraint and the taboo table mechanism are combined, the model can dynamically adjust the generated probability distribution, the more appropriate candidate token can obtain the higher probability, and therefore the accuracy of code completion is improved.

The invention relates to the technical field of federated learning, in particular to a dynamic federated mutual learning method and system for balancing individuation and generalization, and the method specifically comprises the following steps: each client carries out the preprocessing of data to be processed of a model, and carries out the strong enhancement and weak enhancement processing; inputting the data subjected to strong enhancement processing into a shared model, inputting the data subjected to weak enhancement processing into a private model, and performing iterative training on the two models; related parameters of the shared model after each round of iterative training and a difference item between two model parameters are uploaded to a federation server; the federated server adopts a multi-dimensional adaptive aggregation strategy to obtain an updated global model, and returns the updated global model to each client to replace the shared model in the next round of training; and finally generating a generalization result and a personalized result. According to the method, the private-shared model architecture is constructed, and dynamic federated mutual learning is carried out in combination with the federated server, so that balance and collaborative improvement of individuation and generalization performance can be realized.

The invention belongs to the technical field of neural network model security, and discloses a hierarchical encryption neural network model encryption method and system, and the method comprises the steps: obtaining model architecture data, hierarchical topology data and weight parameter data; constructing a model hierarchy sensitivity map for vulnerability analysis to obtain a key horizon map; constructing a parameter importance network for sensitivity analysis to obtain a core parameter set; constructing a model protection strategy knowledge base; performing equipment fingerprint analysis to obtain equipment unique identification data; performing matrix transformation detection to obtain a hierarchical transformation function family; generating a hierarchical key pedigree; performing hierarchical encryption analysis to obtain a hierarchical encryption matrix; performing hierarchical protection conversion by using the hierarchical encryption matrix to obtain a model protection version; performing multi-dimensional integrity verification to obtain a target encryption model; monitoring the operation safety state of the model in real time, and optimizing a hierarchical encryption matrix; and the safety of the model is greatly improved.

The invention relates to an arc fault detection method and system based on a dynamic fuzzy threshold and a storage medium, and the method comprises the steps: collecting the current circuit data of to-be-detected electrical equipment, and carrying out the feature extraction of the current circuit data, and obtaining a current feature parameter; the method comprises the following steps: training historical circuit data by taking a decision tree algorithm as a model architecture, and in the training process, performing parameter updating through a double-sliding window mechanism and optimizing a fuzzy threshold band range through an information gain maximization principle to obtain a dynamic fuzzy threshold decision model; and performing arc fault analysis on the current characteristic parameters through the dynamic fuzzy threshold decision model to obtain a fault detection result, and outputting early warning information when the fault detection result is that an arc fault occurs. According to the method, the dynamic fuzzy threshold value band is constructed through the Gaussian mixture model, and the self-adaptive adjustment of the threshold value is realized in combination with a double-sliding-window online updating mechanism. According to the technical scheme, the accuracy of arc faults can be remarkably improved, and particularly the false alarm rate is reduced.

A computer-implemented method for training a generative adversarial network (GAN) for enhancing underwater images. An adversarial loss is computed for updating a discriminator model and a combined loss is calculated for updating a generator model. The combined loss is calculated based on loss components including the adversarial loss and at least one further loss component. Additionally disclosed herein is a generator network for processing underwater images that includes a novel encoder-decoder model architecture. Unlocking insights from Geo-Data, the present invention further relates to improvements in sustainability and environmental developments: together we create a safe and liveable world.

The invention discloses a hydraulic power plant automatic debugging system with a self-adaptive communication function. The system comprises a processing unit, a storage unit, a communication module, an input / output interface, a task classification module, a model architecture configuration module and other functional modules. The use method comprises the following steps: analyzing task complexity and high-dimensional features of input data through a task classification module, and grading by adopting algorithms such as principal component analysis and a support vector machine; the model architecture configuration module is used for matching a model architecture according to the complexity level and determining an initial calculation path; the dynamic path adjustment module optimizes a calculation path in combination with real-time data, and the resource distribution module redistributes resources when the calculation power exceeds the limit; the model compression module generates a compression model adaptive to the edge device through pruning and quantization algorithms; the parameter fine tuning module is used for jointly optimizing model parameters for a multi-task scene; according to the method, the adaptability and efficiency bottleneck of an existing debugging tool in a complex scene is effectively solved, and the intelligent debugging level of an automatic system of a hydraulic power plant is improved.