17results about How to "Improve model performance" patented technology

The invention provides a target identification method and system based on multi-source information fusion, and relates to the technical field of low-altitude target detection. The method comprises the following steps: acquiring target radar track data, interception equipment track data and position area information data; based on target radar track data, extracting a first feature in an RCS form dimension, and extracting a second feature in a motion dimension; based on the target radar track data and the track data of the monitoring equipment, determining a frequency spectrum monitoring correlation factor and regional position information features; performing feature fusion on the first feature, the second feature, the spectrum interception correlation factor and the regional position information feature to obtain a target feature, and identifying a target type; and identifying a target threat level based on the target type and the target radar track data. The method and the device are used in a target identification process based on multi-source information fusion, and the technical problem that the target threat degree cannot be accurately identified in a complex environment in the prior art is solved.

The invention provides a malicious user identification method and system giving consideration to privacy protection in a social network, and relates to the technical field of network privacy security, and the method comprises the steps: carrying out the structure perception sub-graph segmentation of a social network graph through an METIS algorithm, dividing an original graph into a plurality of sub-graphs, minimizing the number of edges crossing the sub-graphs, and keeping the scale balance of the sub-graphs; constructing a privacy perception GNN of an integrated gating residual attention module, wherein the privacy perception GNN comprises a privacy perception linear layer and a gating residual mechanism; based on differential privacy stochastic gradient descent framework training, combining an adaptive noise scheduling strategy, dynamically adjusting the noise scale according to privacy consumption deviation, and performing closed-loop control budget to obtain a trained model; and malicious users are identified through the trained model. According to the method, the problem of performance reduction caused by fixed noise injection and noise amplification is solved, and efficient and robust identification of malicious users is realized while differential privacy constraints are met.

The application provides a machine translation model generation method and device, comprising: obtaining training data; training a machine translation model based on a pre-stored deep neural network using the training data; the machine translation model is used to translate a text to be translated into a text translation; the machine translation model comprises a memory enhancement adapter layer, a first memory and a second memory; the memory enhancement adapter layer is used to retrieve information in the first memory and / or the second memory and utilize the information; the first memory is used to save source language data obtained by performing reverse translation and forward calculation processing on the training data; and the second memory is used to save target language data obtained by performing reverse translation and forward calculation processing on the training data. The application saves the training data in a form acceptable to the model, reads information from the training data to assist the model when performing a translation task, and achieves the effect of adapting to various translation scenarios while taking into account lower algorithm time complexity and better model performance.

The application provides an end-edge-cloud collaborative hierarchical federated learning training method, device and storage medium, and belongs to the technical field of distributed machine learning and edge computing. The method is applied to a three-layer system including a cloud server, an edge server and a terminal device, and comprises the following steps: the cloud server distributes a global model to each edge server; the edge server dynamically clusters according to the computing and communication capabilities of the subordinate terminal devices, determines differentiated model calculation compression rates and communication compression rates for different clusters, and distributes lightweight models; the terminal device performs local training and uploads parameters; the edge server and the cloud server perform teacher-free online knowledge distillation based on a shared data set to collaboratively update the model; and the cloud server triggers global aggregation according to a dynamic time threshold scheduling strategy to generate a new round of global model. The application effectively alleviates the training blocking problem in a heterogeneous device environment, improves resource utilization efficiency and model performance, and enhances system convergence stability.

The application discloses a double-embedding model hybrid training method, system, device and storage medium, which is applied to the technical field of recommendation systems and comprises the following steps: determining the routing parameters of each classification feature based on the historical access information of the classification features in a training data set; selecting a corresponding embedding table for the classification features input currently according to the routing parameters; wherein the embedding table comprises a first embedding table and a second embedding table; extracting the embedding vector corresponding to the classification features from the selected embedding table, and performing forward calculation and loss calculation of the model based on the embedding vector; and updating the parameters of the first embedding table, the second embedding table and a downstream model according to the result of the loss calculation. The application simulates the embedding hybrid use mode in reasoning in the training stage, enhances the collaborative ability of the two embedding tables, and improves the performance and robustness of the model in a real reasoning scene.

The application relates to the technical field of aerospace and artificial intelligence, and provides a space target orbit intelligent prediction method fusing an orbit dynamics constraint, which comprises the following steps: preprocessing original orbit data of a space target to generate a complete time series data set with continuity and uniform time intervals; constructing a physical information neural network model which represents input and output in the form of time series data, and performing differential calculation on an output sequence by using an automatic differentiation mechanism; designing and introducing a loss function with an adaptive weighting mechanism to dynamically balance loss terms of network modules and physical modules in a training process; training the physical information neural network model; setting a data subset sampling rate, constructing a multi-scale data scene, and evaluating performance of the physical information neural network model under different data scale conditions. The application can improve modeling capability for a long-time orbit evolution process and realize dynamic adjustment of weights of various loss terms.

This invention relates to the field of remote sensing image processing technology, and in particular to a method for extracting urban green spaces from collaborative multi-source remote sensing imagery. The method includes acquiring multi-source remote sensing image data to construct an urban green space information map, preprocessing the images and extracting features to generate image feature vectors, fusing them through semantic mapping to generate a comprehensive feature vector, training a green space extraction model for small-sample scenarios based on a transfer learning framework, and using real-time imagery to determine green space types and update the map. This invention can improve the accuracy of urban green space extraction, adapt to small-sample scenarios, and achieve dynamic monitoring and information improvement.

The invention provides a multi-agent collaborative machine learning process automatic generation method and system, and the method comprises the steps: A1, obtaining a target data set from a code warehouse, obtaining the feature information of the data set, and generating a data analysis result; a2, according to the data analysis result, selecting a corresponding process template as prior information, and then scheduling generation of a plurality of agents and obtaining an agent generation result; step A3, verifying and integrating the agent generation results, and forming and executing a complete machine learning process; and A4, after the machine learning process is successfully executed, searching optimal hyper-parameter configuration and outputting a final model result. According to the method, automatic generation of the end-to-end machine learning process under multi-agent cooperation can be realized, and the success rate of process generation, the model performance and the robustness of the system are remarkably improved.

The invention belongs to the technical field of burn personalized skin scaffold systems, and particularly relates to a burn personalized skin scaffold design method and system fusing 3D printing and an adaptation algorithm, and the method and system carry out multi-modal data synchronous collection, modal noise reduction and standardized safety storage. A high-precision clock module is additionally installed for OCT equipment, a temperature sensor, a pH biosensor and a fluorescence imaging module, unified acquisition software is established to set a corresponding sampling frequency, multi-equipment synchronous acquisition is realized through single triggering, and a fast expansion random tree and a particle swarm optimization algorithm accelerated by a GPU are adopted; an edge calculation module is integrated at a printing equipment end, a real-time data processing framework is adopted, a physical and digital twin system is established to optimize printing parameters to perform closed-loop control, and testing is performed on a bionic skin model containing a dynamic fluid circulation system and a mechanical stretching device.

The invention discloses a code reconstruction method and device, equipment, a medium and a chip, and belongs to the field of artificial intelligence. Comprising the steps of generating a UML class diagram according to a call chain of a source code of a to-be-upgraded application; the UML class diagram is a code class relation diagram meeting the reconstruction requirement of the source code; performing function marking on a plurality of functions contained in the source code to obtain function marking information; generating at least one code reconstruction request according to the source code, the reconstruction demand of the source code and the function mark information; a target code is obtained through a pre-trained code reconstruction model according to the code reconstruction request, and the code reconstruction model is obtained through training based on a GPT large language model: a target code framework is obtained through the pre-trained code reconstruction model according to the UML class diagram and the function marking information; and splicing the target code with the target code framework to obtain a reconstructed code of the source code.

The application discloses a graph learning method, comprising: sampling original space-time information of nodes in a graph to obtain a set of space-time sequences of the nodes in the graph; encoding each node in the set of space-time sequences according to a space-time sequence in the set of space-time sequences to obtain an encoding matrix of each node; obtaining an encoding matrix sequence of each space-time sequence in the set of space-time sequences according to the encoding matrix of each node; and fusing the encoding matrix sequences corresponding to the space-time sequences with the same target node in the set of space-time sequences to obtain space-time attribute information of the target node. The technical scheme provided by the application solves the problem in the prior art that only the space information of nodes is collected in graph learning, and rich data information cannot be provided for downstream specific services, and the model performance of the downstream specific services is improved.

The invention relates to the technical field of federated learning and block chains, in particular to a federated learning incentive method and system based on two-stage commitment and a block chain. According to the technical scheme, the federal learning excitation method based on the two-stage commitment and the block chain comprises the following steps: a reputation evaluation step: dividing interaction behaviors of a client into three conditions of positive interaction, negative interaction and uncertain interaction according to a deviation degree between each round of uploading model updating and a global updating direction of the client, the reputation value of the client is calculated by adopting a subjective logic-based Bayesian reasoning method, and the influence of long-term interaction on the reputation is weakened by introducing a time decay mechanism. By introducing the block chain and the dual-stage commitment mechanism, it is ensured that the server cannot deduct the deserved rewards of the client without being found, the publicity, transparency and fairness of the incentive process are ensured, and the trust of each participant is maintained.

The application belongs to the technical field of artificial intelligence and relates to a data prediction processing method based on artificial intelligence, which comprises the following steps: generating a target inner product based on feature data, a weight coefficient and an inner product; generating a first-order partial derivative and a second-order partial derivative of the target inner product; constructing a diagonal matrix based on the second-order partial derivative, constructing first initial data based on the diagonal matrix and the feature data, and constructing second initial data based on the diagonal matrix, the target inner product and the first-order partial derivative; encrypting the diagonal matrix, the first initial data and the second initial data to obtain encrypted data; calculating an inverse matrix and dividing the encrypted matrix to obtain a matrix; dividing the encrypted matrix to obtain a division matrix based on the encrypted matrix, and generating a first model weight coefficient based on the division matrix; generating a target model weight coefficient based on the first model weight coefficient and a second model weight coefficient; and performing prediction processing on business data based on the target prediction model. The application effectively improves the training efficiency and model effect of the target prediction model.

The invention discloses a privacy protection federated learning method for large-scale pollution data processing. According to the method, small gradient sampling and differential privacy stochastic gradient descent algorithms are combined, a small gradient sampling method is expanded into a machine learning multi-classification model, the feasibility of the small gradient sampling method is proved through theoretical analysis and practical application, then the small gradient sampling method is used for overall data, high-probability unpolluted data in original data are extracted, and the data are extracted to be classified into a multi-classification model. Training a model for the extracted sub-data by using a differential privacy stochastic gradient descent method; the result shows that the differential privacy stochastic gradient descent algorithm improved based on small gradient sampling can perform data analysis and model training on large-scale pollution data, and meanwhile, the resistance and privacy protection effect of federal learning on the large-scale pollution data are enhanced.

This invention provides a model training method, apparatus, device, and storage medium, comprising: acquiring target sample data, determining positive sample data and difficult negative sample data; extracting features from the target sample data, positive sample data, and difficult negative sample data using an initial deep learning model to obtain a first feature vector, a second feature vector, and a third feature vector; determining the similarity between the first and second feature vectors and the similarity between the first and third feature vectors to obtain a third similarity value and a fourth similarity value; determining a first loss value and a second loss value; and training the initial deep learning model using the first and second loss values ​​to obtain a target deep learning model. This application aims to improve the training effect and performance of the model based on positive sample data and difficult negative sample data. Especially for medical image recognition models, it can improve the model's recognition efficiency and accuracy for medical images.

The application provides a big data fast processing and abnormality identification system and method. The system comprises a data acquisition and access module, which can acquire data from heterogeneous data sources in multi-thread parallel mode and standardize the data; a distributed buffering and storage module, which adopts Kafka and HDFS to realize high-throughput buffering and reliable storage; a streaming processing calculation module, which can aggregate features based on a dynamic sliding window and can also calculate trend features; an abnormality identification engine, which is internally provided with an adaptive threshold algorithm model, the tolerance coefficient of which can be dynamically adjusted, and has a cold start strategy for new data streams; an alarm and action execution module, which can trigger multi-level alarms and automatic actions; and a system management and visualization module, which provides configuration, display and other functions. The method covers the whole process from data acquisition to model iteration. The application can efficiently process data, intelligently identify abnormalities, and continuously improve the model performance and abnormality identification accuracy.

This invention relates to an adaptive signal control method and system for intersections based on a large language model. The method includes: acquiring real-time traffic data and preprocessing it, then inputting the data into a traffic signal optimization model to obtain the optimal signal phase. The training process of the traffic signal optimization model includes: acquiring historical traffic data and preprocessing it, then inputting it into a policy network; the policy network outputting the optimal signal phase and obtaining a reward with the goal of optimizing the total additional waiting time; constructing an experience buffer based on historical traffic data and rewards, and having a value network sample from the experience buffer to calculate the state value, advantage function, and total loss function, and updating the parameters of the policy network and value network until training converges, resulting in a trained traffic signal optimization model. Compared with existing technologies, this invention reduces the training cost of the model while ensuring the intersection control effect, and enhances the interpretability and generalization ability of signal control decisions.