Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

45 results about "Kernel method" patented technology

In machine learning, kernel methods are a class of algorithms for pattern analysis, whose best known member is the support vector machine (SVM). The general task of pattern analysis is to find and study general types of relations (for example clusters, rankings, principal components, correlations, classifications) in datasets. For many algorithms that solve these tasks, the data in raw representation have to be explicitly transformed into feature vector representations via a user-specified feature map: in contrast, kernel methods require only a user-specified kernel, i.e., a similarity function over pairs of data points in raw representation.

Point projection type three-dimensional reconstruction and segmentation method and system based on semi-Gaussian pruning

The invention relates to the technical field of computer vision, in particular to a point projection type three-dimensional reconstruction and segmentation method and system based on semi-Gaussian pruning. The method comprises the following steps: respectively obtaining an SFM point cloud and a consistency label mask of a cross-view label based on an obtained multi-view image; initializing the obtained SFM point cloud into an identity semi-Gaussian point cloud, and performing rendering optimization by using a differentiable renderer; densifying the initial sparse point cloud by using a localized semi-Gaussian point management method, and identifying a local error region for resetting and repairing; using the obtained consistency label mask to supervise Gaussian identity feature learning by using cross entropy loss, and using unsupervised 3D regularization loss to force spatially adjacent gauss to maintain identity consistency; according to the method, the identity coding semi-Gaussian kernel method is adopted, the inherent representation fuzziness of a single opacity formula is eliminated, and the positive influence on the identity coding precision is generated.
Owner:YANTAI UNIV

Multi-mode fusion sea surface target identification method of high-dynamic unmanned platform

The invention relates to the technical field of sea surface target multi-mode fusion recognition, in particular to a multi-mode fusion sea surface target recognition method of a high-dynamic unmanned platform. According to the method, a multi-modal fusion recognition framework based on a multi-kernel learning method is provided, and a small-sample multi-modal fusion target recognition algorithm for a sea-surface target is designed on the basis of the multi-modal fusion recognition framework. Target detection and feature extraction are respectively performed according to data characteristics of different sources of visible light images, SAR images and active and passive radar one-dimensional range profile signals, multi-mode heterogeneous features are mapped by using a multi-scale kernel method, weighted fusion of a kernel matrix is performed, global training optimization is performed, and finally a classification result of each sample can be output. The algorithm not only considers the performance of a data-driven target recognition algorithm, but also considers platform hardware computing power limitation and small sample conditions, so that the balance among the performance, computing power and samples can be realized.
Owner:NORTHWESTERN POLYTECHNICAL UNIV

Calculation method and device for multi-scale particle multiphase flow field and electronic equipment

The invention discloses a calculation method and device for a multiphase flow field of multi-scale particles and electronic equipment, and the method comprises the steps: determining large-scale analytic particles and small-scale non-analytic particles in a multiphase flow according to a preset computational domain grid size; acquiring flow field medium parameters of the analytic particles and physical parameters of discrete particles, and solving flow field information and physical information of multiphase flow by using an immersion boundary method; interface parameters of each medium composition in the non-analytic particle flow field and physical parameters of discrete particles are obtained, the particle volume fraction is solved through a statistical kernel method, and based on the particle volume fraction, physical information of multiphase flow is solved through a fluid volume method; and according to the physical information solved by the analytic particles and the non-analytic particles, calculating the physical parameters of the multi-scale particle multiphase flow field.
Owner:HEFEI GENERAL MACHINERY RES INST +2

Semi-supervised multi-view clustering method based on double-constraint concept decomposition

ActiveCN120995142AMachine learningKernel methodEngineering
The invention discloses a semi-supervised multi-view clustering method based on double-constraint concept decomposition, which relates to the technical field of machine learning, combines point constraint with pairwise constraint, constructs a double-constraint algorithm based on prior label information of a complementary supervision mechanism, and realizes maximum utilization and progressive propagation of limited supervision information. And based on a concept decomposition framework, breaking through the limitation of a traditional method on non-negativity of data, processing complex data distribution in combination with a kernel method, obtaining a consensus matrix of multi-view data fusion as low-dimensional representation of views, performing clustering by adopting a k-means algorithm, and outputting a final category division result. According to the method, effective processing of complex multi-view data is realized, finally, the performance of a multi-view clustering task is comprehensively improved, and the bottlenecks in clustering precision, data adaptability and supervision information utilization efficiency in the prior art are broken through.
Owner:GUANGDONG UNIV OF TECH

Quantum computing with kernel methods for machine learning

Methods, systems, and apparatus for quantum machine learning. In one aspect, a method includes obtaining, by a quantum computing device, a training dataset of quantum data points; computing, by the quantum computing device, a kernel matrix that represents a similarity between the quantum data points included in the training dataset, comprising computing a value of a kernel function for each pair of quantum data points in the training dataset, wherein the kernel function is based on reduced density matrices for the quantum data points; and providing, by the quantum computing device, the kernel matrix to a classical processor, wherein the classical processor performs a training algorithm using the kernel matrix to construct a machine learning model.
Owner:GOOGLE LLC

Federal learning method and system based on core, computer equipment and storage medium

PendingCN120338141AKernel methodsDigital data protectionAlgorithmKernel method
The invention relates to a kernel-based federal learning method and system, computer equipment and a storage medium. Target clients are selected from a client set to form a client subset; sending a first parameter of the first machine learning model and the number of iterations to the client subset; wherein the target client is used for performing random feature sampling and double stochastic gradient evaluation according to the first parameter and the number of iterations, and updating a second machine learning model based on double stochastic gradients to generate a second parameter; the first machine learning model is a local machine learning model of the server, and the second machine learning model is a local machine learning model of the target client; receiving a second parameter returned by the client subset, and updating the first machine learning model according to the second parameter; the expansibility of the kernel method in a federated learning scene is improved; and model convergence acceleration is facilitated.
Owner:HANGZHOU HIGH-TECH ZONE (BINJIANG) INSTITUTE OF BLOCKCHAIN & DATA SECURITY +1

Renewable energy data clustering method and system based on optical quantum computer

The invention discloses a renewable energy source data clustering method and system based on an optical quantum computer, and the method comprises the steps: building a renewable energy source data clustering optimization problem based on a quadratic unconstrained binary optimization (QUBO) model based on a discrete optimization objective function, carrying out the matrix of the objective function, carrying out the clustering optimization of the objective function through a kernel method, and carrying out the clustering optimization of the objective function. The method comprises the following steps of: establishing a target function item, converting the target function item into a matrix form, constructing a matrix of a new clustering model, finally solving a new QUBO optimization problem by using CIM, and converting the new QUBO optimization problem into an optimal clustering result. According to the method, data similarity measurement is improved by constructing the QUBO model and combining a kernel method, and a renewable energy source data clustering problem is converted into a form suitable for CIM solution, so that the calculation bottleneck of a classical calculation method in high-dimensional data processing is avoided. According to the method, the nonlinear characteristic of the data can be kept, efficient clustering optimization calculation can be realized, and the calculation efficiency and the clustering precision are remarkably improved.
Owner:SOUTHEAST UNIV

Nuclear subspace optimization method for full-wave inverse scattering high-precision imaging

The invention discloses a nuclear subspace optimization method for full-wave inverse scattering high-precision imaging, and relates to the field of machine learning and full-wave inverse scattering in order to solve the problem that an existing subspace-based optimization method is insufficient in precision in the field of full-wave inverse scattering imaging. According to the method, the robustness framework of a traditional subspace optimization method and the self-adaptive learning advantage of a kernel method are combined, iteration of an optimization method of an original subspace is transferred into a regeneration kernel Hilbert space, and evolution in the iteration process is learned with the minimum mean square error as the learning criterion. According to the kernel-based subspace optimization method, a mature and interpretable framework of a subspace optimization method is combined with the adaptive learning ability of a kernel method, a robust and universal algorithm is provided for full-wave inverse scattering in the fields of medical imaging, geophysical exploration, nondestructive testing and the like, and an accurate reconstruction result can be provided.
Owner:UNIV OF ELECTRONICS SCI & TECH OF CHINA

A Drought Prediction Method Based on Energy Flux, Causal Analysis, and Machine Learning

This invention discloses a drought early warning method based on energy flux, causal relationship analysis, and machine learning. The method includes the following steps: S1, acquiring energy flux and drought indicators, determining the optimal lag time through causal reasoning, and constructing a multi-order lag feature set; S2, establishing a tree model, using regression / kernel methods and a time series model candidate set, optimizing hyperparameters through particle swarm optimization, integrating two layers in a stacked manner, adaptively optimizing the performance of comprehensive regression and event recognition, and outputting a predicted sequence; S3, setting multi-level early warning rules according to drought thresholds, mapping drought levels, and evaluating effectiveness through statistical precision and recall; S4, calculating contribution using an additive feature attribution algorithm, identifying nonlinear thresholds to form sensitivity analysis, and improving interpretability. This invention achieves a 7-11 month early warning prediction of drought based on energy flux, with a drought early warning recall rate of 66.67%-75.86%, significantly improving the accuracy and interpretability of drought early warning.
Owner:CHINA INST OF WATER RESOURCES & HYDROPOWER RES

Mendel randomization nonlinear causal inference method based on genome research

The invention provides a Mendel randomization nonlinear causal inference method based on genome research, and relates to the technical field of biological genetic data analysis, and the method comprises the following steps: obtaining a tool variable, an exposure factor and a result variable, and carrying out data preprocessing to form a structured input matrix; respectively constructing a tool variable nonlinear model and a causal effect nonlinear model; calculating a loss function of the model, and adding a regularization item for joint optimization; optimizing model parameters by using a GBDT kernel method; updating the model based on the optimized model parameters and drawing a nonlinear response curve of the exposure factors and the result variables; replacing the tool variable with a pseudo tool variable to repeat the above steps to verify a causal inference result, and inferring a causal effect according to the verified causal inference result; according to the method, the limitation of assuming a linear relation and processing high-dimensional data in a traditional causal inference method can be overcome, and a more accurate, robust and interpretable causal inference result is provided.
Owner:GANSU WANWEI INFORMATION TECH CO LTD

Distributed MIMO radar AOA target positioning method based on kernel recursive least square

The invention discloses a distributed MIMO radar AOA target positioning method based on kernel recursive least squares, and the method comprises the steps: carrying out the nonlinear mapping of AOA measurement data obtained by a radar to a high-dimensional RKHS based on a kernel method, converting a nonlinear target position estimation problem into a linear parameter estimation problem in the RKHS, carrying out the online learning through employing the obtained AOA measurement data, and carrying out the positioning of a target position. And outputting a position estimation result of the target on line to complete target positioning. According to the method, effective target positioning precision can be provided under the condition that measurement error prior information is unknown, target positioning precision higher than that of an existing method can be obtained under the condition that measurement errors are large, and the problem that timeliness still needs to be guaranteed under the condition that the requirement for radar target positioning precision is high is solved.
Owner:UNIV OF ELECTRONICS SCI & TECH OF CHINA

A new method for identifying RNA pseudouridine sites

This solution discloses a new method for identifying RNA pseudouridine sites. This method proposes to use a variety of feature representation techniques to extract sequence features, and then uses the SVM-RFE method for feature selection to compress the feature space and optimize the feature subset. The best feature set after feature selection is input into the kernel method KeMRF based on polynomial random forest to identify pseudouridine sites in the sequence. As a newly proposed classification method, compared with the traditional random forest, KeMRF not only optimizes the discriminant criterion for node splitting, but also combines with an easy-to-interpret kernel method, making the classification performance more superior. This method reduces the training time of the model, improves the classification performance of the model, and further enhances the accuracy of identifying pseudouridine sites.
Owner:YANGTZE DELTA REGION INST (QUZHOU) UNIV OF ELECTRONIC SCI & TECH OF CHINA

Online adaptive kernel selection method for image multi-label data stream classification

The invention discloses an online adaptive kernel selection method for image multi-label data stream classification in the technical field of machine learning. The method comprises the following steps: a program receives a feature vector of a to-be-identified image; respectively mapping the feature vectors to a feature space by using a feature mapping subprogram to obtain mapped feature vectors; using a current latest multi-label kernel classifier to predict a related label set of the feature vector in real time; after the prediction is finished, an online updating subprogram of the multi-label kernel classifier is entered, and the subprogram obtains an iterative updating formula of the multi-label kernel classifier, so that the multi-label kernel classifier of the next round is obtained; waiting for the next round of image feature vector; according to the method, kernel function selection can be executed in the online learning process, meanwhile, online learning of the multi-label kernel classifier is carried out, and the online multi-label prediction performance of an existing online kernel method is greatly improved.
Owner:YANGZHOU UNIV

Image determination method and device, equipment, medium and product

The invention discloses an image determination method and device, equipment, a medium and a product, the method is applied to a multi-core processor, the multi-core processor at least comprises two cores, and the method comprises the following steps: obtaining radar echo data and the number of cores included in the multi-core processor; distributing the radar echo data according to the number of the kernels, the total number of the data and the number of preset butterfly levels, and determining a target data range processed by each kernel; and through each kernel, performing Fourier transform butterfly operation on the target data in the target data range according to a preset butterfly level to obtain a two-dimensional image. And processing data of each kernel is reasonably distributed according to the total data quantity, the kernel quantity and the preset butterfly level quantity, and a plurality of kernels perform butterfly operation and inter-kernel rapid data transmission in parallel to obtain a two-dimensional image. Each kernel reaches the optimal calculation state through data distribution, the overall calculation rate is improved, the number of the kernels can be configured, and the universality and ductility of the method are improved.
Owner:SHANGHAI SMARTLOGIC TECHNOLOGY LTD

A relation extraction method based on neural kernel method

ActiveCN118940746Bimprove performanceEffective decision boundary learningSemantic analysisNeural learning methodsKernel methodRelationship extraction
A relation extraction method based on neural kernels, belonging to the field of natural language processing, includes: defining and formalizing the relation extraction task, establishing a mapping from a set of relation instances to a set of relation labels; constructing a neural kernel model, assigning input data to the same type as the closest labeled reference instance. This invention introduces three neural kernels to demonstrate the kernel substitution mechanism in relation extraction: by optimizing the three kernels and constructing a composite kernel using kernel substitution, the discriminative power of the deep neural network is improved. This invention utilizes neural kernels for relation extraction, improving classification accuracy by combining neural networks and kernel substitution. The neural kernel method can automatically learn the similarity between relation instances, avoiding the problems that may arise from manually designing features and distance functions in existing relation extraction methods, while fully utilizing deep learning and external resources to improve the performance and effectiveness of the relation extraction task.
Owner:GUIZHOU UNIV

Big data integration task management system based on cloud edge collaboration

The invention discloses a big data integration task management system based on cloud edge collaboration, and the system comprises the following modules: a data feature template library construction module which is used for generating a data feature template library and forming a grassland nutrition degree score set; the goat foraging optimization algorithm candidate task scheduling graph population initialization module is used for dividing a hillside memory area, a valley bottom memory area and a ridge memory area according to node aggregation degrees; the nerve tangent kernel method performance prediction module is used for fusing into a nerve tangent kernel method performance prediction vector; the fitness function construction and optimization module is used for executing global exploration and local development of a goat foraging optimization algorithm population and outputting an iterative optimization result; and executing a directed acyclic graph generation and deployment module for completing initial deployment of a multi-source heterogeneous data source integration task. According to the method, the resource utilization efficiency is maximized, seamless switching among the three mainstream computing engines is realized, and the deployment scheme is elastically adjusted according to the resource load.
Owner:JIANGYIN XINGCHENG TECHNOLOGY ENGINEERING CO LTD

Semi-supervised multi-view clustering method based on concept decomposition of double constraints

ActiveCN120995142BMachine learningKernel methodEngineering
The application discloses a semi-supervised multi-view clustering method based on double-constraint concept decomposition, relates to the technical field of machine learning, combines point constraints with pair constraints, constructs a double-constraint algorithm based on prior label information of a complementary supervision mechanism, realizes maximum utilization and progressive propagation of limited supervision information, breaks through the limitation of data non-negativity of a traditional method by relying on a concept decomposition framework, processes complex data distribution by combining a kernel method, obtains a consensus matrix of multi-view data fusion as low-dimensional representation of a view, adopts a k-means algorithm for clustering, and outputs a final class division result. The application realizes effective processing of complex multi-view data, finally comprehensively improves the performance of a multi-view clustering task, and breaks through the bottleneck of the prior art in clustering precision, data adaptability and supervision information utilization efficiency.
Owner:GUANGDONG UNIV OF TECH

Data missing oriented random workflow load prediction method

This invention discloses a method for predicting the load of stochastic workflows with missing data. The method includes: a workflow preprocessing stage, where the workflow is transformed into an embedded vector form and clustered based on historical data similarity; a workflow missing value completion and load prediction stage, where a kernel density missing data filling method based on common features is used to fill in missing data, and a workflow load analysis and prediction method is proposed based on the characteristics of the computing network-cloud-edge-device structure and historical information on workflow computation, storage, and communication; a cluster modeling and feature engineering stage; and a cluster load prediction stage, where this invention combines kernel methods and Kalman filtering techniques to predict missing values ​​and workflow load, optimizing the robustness and generalization of the prediction. This method has broad application value and promising prospects in the fields of load prediction and workflow scheduling.
Owner:SOUTHEAST UNIV

Systems and methods for quantum circuit simulation using tensor networks

PendingUS20260099752A1Quantum computersMachine learningQuantum circuitKernel method
Embodiments of the present disclosure provide functionality to tensor network framework designed for quantum kernel methods and demonstration of tensor network effectiveness at scaling this application. Quantum kernels capture the distance between data points in quantum feature space by evaluating the quantum state overlaps associated with each data point. It has been found that expressing data in quantum feature space may produce more separable data that improves the results of linear classifiers. The different kernel elements may be computed independently, and parallel processing may be exploited to significantly reduce computational time, enabling to train on more data. Thus, quantum kernels continue to improve classification metrics with the addition of more training data and more features.
Owner:HSBC TECHNOLOGY & SERVICES USA INC

Quantum computing with kernel methods for machine learning

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for quantum machine learning. In one aspect, the method includes obtaining, by a quantum computing device, a training dataset of quantum data points; computing, by the quantum computing device, a kernel matrix representing a similarity between quantum data points included in the training dataset, including computing a value of a kernel function for each pair of quantum data points in the training dataset, wherein the kernel function is based on a reduced density matrix of the quantum data points; and providing, by the quantum computing device, the kernel matrix to a classical processor, wherein the classical processor uses the kernel matrix to perform a training algorithm to construct a machine learning model.
Owner:GOOGLE LLC

Electromagnetic full-wave inverse scattering imaging method based on shape prior information

The invention provides an electromagnetic full-wave inverse scattering imaging method based on shape prior information, and relates to the field of machine learning and electromagnetic full-wave inverse scattering. Aiming at the problem that shape prior, nonlinear modeling and physical interpretability are difficult to effectively integrate and realize in the field of electromagnetic full-wave inverse scattering imaging, the invention integrates shape prior, physical consistency and a kernel method, and relates to the field of machine learning and electromagnetic full-wave inverse scattering. According to the method, singular value decomposition (SVD) is utilized, a low-dimensional shape parameter is adopted to carry out continuous microparameter modeling on a target boundary, and a high-dimensional pixel space is mapped to a structured prior space. Physical constraints are enforced to ensure consistency between state variables and measurement data. In addition, a kernel method is introduced to carry out nonlinear modeling on state-residual mapping, and a dual-time-scale random optimization strategy is adopted to balance parameter flexibility and physical interpretability. According to the method, excellent real-time reconstruction performance is realized in a high-noise and complex-structure scene, and a new stable and explainable method is provided for electromagnetic full-wave inverse scattering imaging.
Owner:UNIV OF ELECTRONICS SCI & TECH OF CHINA

Quantum kernel method, classification method and related systems, devices

ActiveCN117273157BQuantum computersQuantum circuitKernel method
The application provides a quantum kernel method, a classification method, a data coding method and related systems and devices, which are applied to a quantum computer or a quantum simulator. The method comprises: encoding any two m-dimensional data in a plurality of m-dimensional data into n quantum bits through a quantum circuit corresponding to a preset kernel function, m and n are positive integers, and m>n; measuring the encoded n quantum bits to obtain quantum kernel inner products of the two encoded m-dimensional data in a Hilbert space, thereby obtaining a quantum kernel matrix corresponding to the plurality of m-dimensional data. In the application, multi-dimensional data is encoded into a smaller number of quantum bits without compressing data feature information as much as possible, the calculation of the quantum kernel method is completed, the running time and the memory space are saved through the compression of the number of bits, and the calculation efficiency is improved.
Owner:ORIGIN QUANTUM COMPUTING TECH (HEFEI) CO LTD

Method and system for constructing CCNoV data set and WESVM classifier to assist Web program design

PendingCN120669967AProgram documentationIntelligent editorsData setKernel method
According to the method, the debugging problem in Web program design is summarized, and particularly the complex ErrorMsg problem is solved. Based on a code annotation method and a NoValidate attribute method, the ErrorMsg problem is solved, features are designed, a CCNoV data set is constructed, and dimension reduction processing is carried out through PCA and tSNE technologies. Experimental results show that the original 15-dimensional data set is excellent in performance in the WESVM classifier, and the training and testing accuracy is up to 97% or above. However, when the data is reduced to two dimensions, the classification effect of the Gaussian Kernel method is better than that of the Learar Kernel, and the accuracy rate reaches 100%. The method disclosed by the invention not only discloses the advantages of a kernel method in processing dimension reduction data, but also shows that the CCNoV data set and the WESVM classifier constructed by the method can improve the efficiency and the accuracy of Web application program development, and support is provided for a programmer to develop the Web application program.
Owner:HANJIANG NORMAL UNIV

A method for rapid detection of seed germination rate

The application provides a method for rapidly detecting seed germination rate, comprising: obtaining a plurality of seed samples with known germination rates, performing volatile organic compound detection on each sample to obtain volatile component data, and after data cleaning and arrangement, constructing a model of a machine learning algorithm of an integrated model, a linear model and a kernel method; using standardized feature data to establish an optimal prediction model of seed germination rate, which can be used for testing of unknown seed germination rate samples. Compared with a traditional manual statistical detection method, the detection period of the traditional method is compressed to within 2 hours, the detection efficiency is greatly improved, the manual germination and grain-by-grain counting operations are omitted, the labor input is significantly reduced, the human subjective error is reduced, the objectivity and repeatability of the detection result are greatly improved, and the error with the actual measurement value is within ± 5%. The method can be widely applied to seed quality detection scenes of various crops, and has good industry application and promotion value.
Owner:LANZHOU UNIV

Soft-sensing modeling method for multiphase batch processes based on improved geodesic flow core

This invention discloses a soft measurement modeling method for multiphase intermittent processes based on an improved geodesic manifold kernel. The method includes: collecting platform parameters from a cloud server; projecting source and target domain data onto a common manifold subspace using a geodesic manifold kernel method based on linear local tangent space arrangement to reduce the data distribution differences between the source and target domains; using a time-series-based fuzzy clustering method to divide the source domain data into phases, obtaining phase division points; and establishing sub-phase soft measurement models based on different phase characteristics using different modeling methods. For phases with slowly changing process characteristics, partial least squares regression is used to establish the soft measurement model; for phases with rapidly changing process characteristics, partial least squares regression based on real-time learning is used to establish the soft measurement model. This method establishes sub-phase soft measurement models for multiphase intermittent processes, enabling the estimation of difficult-to-measure variables and improving the performance of the soft measurement model under varying operating conditions.
Owner:JIANGNAN UNIV

Quantum computer, computer system and control method thereof, and control program

PCT designated stageWO2025238999A1Quantum computersKernel methodsFeature vectorKernel method
According to the present invention, the estimation accuracy of machine learning by a quantum kernel method is improved. A quantum computer (3) estimates a kernel function k(xk,xl) of two feature vectors xk, xl (xk, xl have n elements in total, and n is an integer of 2 or more) using the two feature vectors xk, xl and phase parameters φp ,q(x) (xp and xq are the p-th and q-th elements, respectively, in one of the two feature vectors xk, xl). The phase parameters φp, q(x) are calculated by φp, q(x)=(π-αpxp)×(π-αqxq) using adjustment parameters αp and αq.
Owner:SOFTBANK CORPORATION

Method for computer-aided adjustment of the configuration of different textile productions

ActiveCN118147796BKernel methodProcess engineering
The invention relates to a method for computer-aided adjustment of a configuration of different textile productions, wherein production specifications are changed for different productions and a textile material is processed in at least one production step by at least one machine based on the production specifications, the method comprising the following steps: - starting the implementation of at least one or more production experiments, in which at least one production step is implemented with a setting configured and based on production specifications, - determining the influence of the configuration based on the setting in the respective production experiment, - adjusting the setting based on the determined influence in order to determine an optimized setting in terms of production specifications, - starting the implementation of a textile production, in which at least one production step is implemented with the optimized setting and based on production specifications, wherein at least one machine learning method is used in order to implement the adjustment based on the determined influence, wherein the at least one machine learning method is implemented as at least one kernel method.
Owner:SAURER SPINNING SOLUTIONS GMBH & CO KG

Hyperspectral image classification method of semantic-guided kernel low-rank sparse preserving projection

PendingCN121937762AImprove class separabilityOvercoming the limitations of fixed neighborhood representationClimate change adaptationCharacter and pattern recognitionImaging processingKernel method
The invention belongs to the technical field of image processing, and particularly relates to a hyper-spectral image classification method for semantic-guided kernel low-rank sparse preserving projection, which comprises the following steps: firstly, acquiring a training sample set and a test sample set of a hyper-spectral image, and calculating a kernel matrix and a cross kernel matrix of the hyper-spectral image; further constructing a kernel method dimension reduction model objective function fusing a sparse reconstruction error term, a sparse regularization term, a semantic guidance low-rank constraint term and a spatial adaptive manifold regularization term; the complex objective function is efficiently solved by adopting an alternating direction multiplier method, and an optimal projection matrix is finally obtained by introducing an auxiliary variable and alternately updating a projection matrix coefficient, a sparse coding matrix and other variables; and mapping the test sample to a low-dimensional feature space by using the projection matrix so as to finish classification. According to the method, through semantic guidance and spatial-spectral information adaptive fusion, the discrimination ability and classification precision of dimension reduction features are significantly improved.
Owner:ANHUI UNIV OF SCI & TECH

Deep learning model optimization method and system based on semi-structured pruning

The invention discloses a deep learning model optimization method and system based on semi-structured pruning, and relates to the technical field of model optimization, and the method comprises the following steps: obtaining a yolov5s neural network model trained based on a training set, and marking the model as an initial training model; generating a father-child layer grouping list based on the dependency relationship between the father-child layers of the initial training model; pruning the initial training model based on all pruning convolution kernel methods to obtain an initial pruning mode, and screening the initial pruning mode based on a father-child layer grouping list and a calculated L2 norm to obtain a final pruning model; performing training adjustment on a non-zero weight part of the final pruning model based on the training set; the method is used for solving the problem that in an existing semi-structured pruning deep learning model optimization technology, layer-by-layer analysis and configuration are needed, and consequently calculation complexity is high.
Owner:BEIJING ZHONGKE STRON CLOUD INTELLIGENT TECH CO LTD