18 results about "Evolution strategy" patented technology

This application provides a recommendation method, apparatus, and computer-readable storage medium based on multi-objective optimization. The method includes: acquiring user rating information, including ratings from multiple users for multiple items; clustering users into different clusters using K-means clustering based on the user rating information; predicting the predicted ratings of users for items in each cluster based on a probability propagation algorithm with improved resource allocation; generating an initial population in each cluster based on the predicted ratings of users for items; and solving a multi-objective optimization problem based on an accuracy objective function and a diversity objective function using an evolutionary strategy guided by accuracy-preference users, thereby obtaining the target recommendation result for each cluster. By using the above method to guide the evolutionary direction of the algorithm through user preferences, the target recommendation result can be made more biased towards the accuracy objective function while taking into account diversity.

This invention discloses a knowledge graph-based intelligent auxiliary decision-making generation method for railway stations, comprising the following steps: S1, collecting multi-source operational data of the railway station and preprocessing it; S2, establishing a node set according to preset entity mapping rules and establishing an edge set based on semantic decision relationships to generate a knowledge graph; S3, generating vector representations of each node in the knowledge graph, selecting a starting node, and performing structural traversal based on multi-hop adjacency relationships to establish a state subgraph; S4, performing strategy path evolution in the state subgraph and modeling the conflict relationships between path sequences; S5, parsing the task dependencies in the paths, assembling node content and associated attributes, generating and issuing auxiliary decision-making instructions; S6, collecting feedback data and incrementally updating node attributes and path evolution strategies. This invention enables the structured generation of auxiliary decision-making instructions for railway stations, improving the stability of the decision-making process.

The application relates to the technical field of talent evaluation, and discloses a multi-dimensional adaptive examination generation method and system based on a dynamic skill map. The method comprises the following steps: real-time analysis of recruitment website and enterprise job description data, extraction of atomized skill phrases containing skill names, levels and correlation degrees, and generation of unified semantic representation; obtaining a standardized skill map through a large language model and transmitting the standardized skill map to a central database through a blockchain; generating an evolution strategy based on skill node weights and user historical answering characteristics, and differentiating and executing the evolution strategy by edge nodes; selecting a target edge node and configuring a skill knowledge cache; analyzing a candidate natural language answering request to generate ability matching characteristics, determining a candidate test question set, and outputting a test paper; and updating an ability evaluation model and an evolution strategy in combination with an answering result and a response time. The application can improve the authority and reliability of the evaluation result to meet the scientific and accurate needs of enterprises for talent evaluation.

The invention belongs to the technical field of computer vision, and particularly relates to an end-to-end shape recognition method and system based on a complex network. The method comprises the following steps: carrying out dense discretization on the edge of a to-be-identified image to obtain an edge description point set; selecting a series of key points on the obtained edge description point set at equal intervals; extracting local features of a key point surrounding edge description point set by using a shape context; taking each key point as a node, defining edges in the network according to spatial correlation among the key points, and constructing a multi-layer complex network by adopting a dynamic evolution strategy; local features and topological features of the complex network are processed through a graph convolutional network, and robust expression of the shape classification task is obtained; and performing classification through a full connection layer network to obtain a classification result, and realizing end-to-end shape recognition. According to the method, the adaptability and robustness of the method to different shape classification tasks are enhanced.

The application provides a steel product quality prediction method based on evolutionary strategy-based neural architecture search, and the method comprises the following steps: acquiring historical data and related parameters affecting steel quality under a target production line or scene to construct a feature set required for model training and verification; based on the feature set, a search space containing network structure and hyperparameters is constructed, and a variable-length coding strategy is adopted to encode network layer number, neuron number, activation function, optimizer type and training parameters to generate an initial candidate model population; an improved ES algorithm is adopted to automatically search the structure and hyperparameters of the neural network, and the performance of each generation of candidate models is evaluated by using back propagation training; and the trained optimal model is used for quality prediction of steel production data. The application improves the accuracy of steel product quality prediction.

The invention relates to a DC GIS insulator contour optimization method considering transient reliability. The DC GIS insulator contour optimization method comprises the following steps: defining insulator contour variables based on parameterized curve fitting; calculating a direct current charging electric field based on a partial differential equation; calculating an electric field under the action of tolerable transient voltage based on a partial differential equation; carrying out target optimization modeling; and carrying out optimization solution based on a self-adaptive covariance evolutionary strategy. According to the method, an insulator contour target optimization model taking a basin-type insulator contour variable as a decision variable, taking the weighted sum of the maximum value of the insulator surface electric field intensity under the action of a minimized direct current voltage as a target function, and taking the maximum value of the electric field intensity under the action of polarity inversion and lightning impulse voltage as a constraint condition is adopted; the optimal geometrical shape of the basin-type insulator is solved by utilizing a self-adaptive covariance evolutionary strategy, and the surface electric field intensity of the insulator is reduced to the greatest extent and the insulating property of the insulator is improved on the premise of meeting an allowable value specified by a design criterion and ensuring the transient voltage tolerance reliability.

The application discloses an engineering building structure evolution method, relates to the field of civil engineering and building informatization technology, and realizes dynamic mapping with the entity by constructing a full-element digital twin of the building structure, completes multi-dimensional evolution demand quantitative evaluation, generates a hierarchical evolution strategy and completes intelligent comparison and selection of multiple schemes, verifies the full performance through digital simulation, accurately controls the construction in a digital manner, and finally realizes closed-loop verification of the evolution effect and iteration update of the twin. The application solves the problems that the existing static design of the building structure cannot adapt to changes in the whole life cycle demand, is fragmented in reconstruction, has information faults, has high carbon emission, and lacks hierarchical adaptive evolution strategies, can realize sustainable evolution of the building structure in multiple rounds of the whole life cycle with low disturbance and low carbon emission, greatly reduces the reconstruction carbon emission and the whole life cycle cost, and can be widely applied to sustainable reconstruction of various buildings, municipal and water conservancy structures.

This invention belongs to the field of artificial intelligence algorithm technology and discloses an intelligent transfer adversarial method and system for electromagnetic signal recognition under incomplete information. First, multiple models of alternative target networks with incomplete information are constructed in parallel, and these alternative models are trained using an evolutionary strategy based on a prior knowledge matrix. Then, an ensemble network model is generated based on the trained alternative models, and a fully connected layer and a layer with feature-level weighted fusion are designed. Finally, an optimized adversarial method for electromagnetic spatial signal recognition tasks is designed, generating adversarial samples under the ensemble model, and transferring these adversarial samples to the target network to achieve adversarial capabilities against the target network. This invention can effectively achieve adversarial capabilities against the target network when the target network information is unknown, and it still exhibits good transfer adversarial performance under the same signal recognition task and different unknown target network conditions.

The application discloses a water quality clustering method based on an improved snow melting algorithm, utilizes the improved snow melting optimization algorithm to cluster water quality, finds the optimal division of clustering by using a heuristic algorithm according to a clustering evaluation index, and analyzes and mines multiple groups of related water quality physicochemical indexes covered by water quality data by introducing Latin hypercube sampling to initialize a population, a positive greedy strategy, a self-adaptive differential evolution strategy based on population division and a reverse learning strategy based on average centroid to improve the ability of the method in the clustering problem. The method can not only improve the division precision of water quality, well divide drinkable water and non-drinkable water with different characteristics, realize unsupervised learning of water quality, and has good interpretability.

The invention discloses a class II U-shaped man-machine cooperation disassembly line design method considering the work efficiency risk. The method comprises the steps that disassembly information of a to-be-disassembled mechanical and electrical product is determined; based on the work flow and the disassembly information of the U-shaped man-machine cooperation disassembly line, a disassembly constraint condition is set; based on the disassembly constraint condition, establishing a multi-target U-shaped man-machine cooperation disassembly line task planning model with the minimum beat time, the human body efficiency risk balance index, the disassembly total energy consumption and the disassembly total cost as targets, and converting the multi-target U-shaped man-machine cooperation disassembly line task planning model into a standard mixed integer linear planning model; and constructing a multi-objective evolutionary algorithm based on decomposition and fused with Q learning to solve the model by constructing a coding and decoding strategy based on problem characteristics and an evolutionary strategy and a local search strategy conforming to a decomposition priority relationship, so as to obtain a task planning scheme. According to the method, the ergonomic risk is introduced and quantified, so that efficient collaborative optimization of the beat time, the ergonomic risk balance degree, the disassembly energy consumption and the cost is realized.

This application discloses an equalizer parameter setting method, an audio system, a device, and a readable storage medium. A computing device receives an initial equalizer population; and iteratively selects a first equalizer curve and a second equalizer curve from the initial equalizer population. An audio device selects, based on user preference data, the first equalizer curve or the second equalizer curve as a target equalizer curve. The computing device determines an iterative equalizer population based on the target equalizer curve to acquire an iterative target equalizer curve. After determining a number of iterations reaches a preset value, the computing device determines the final iterative target equalizer curve, and causes the audio device to set the corresponding equalizer parameters based on the final iterative target equalizer curve. This application iteratively selects based on the initial equalizer population, and utilizes a group evolution strategy to achieve a reduction in the search space range, improve convergence speed, and enhance the accuracy of matching between final set equalizer parameters and user preference data.

The application provides an energy-driven unmanned aerial vehicle trajectory optimization method and system based on an evolutionary strategy, and relates to the technical field of unmanned aerial vehicles. The method comprises the following steps: defining total energy consumption and constraint violation functions of an unmanned aerial vehicle according to flight parameters of the unmanned aerial vehicle; constructing a fitness function according to the total energy consumption and the constraint violation functions; initializing a population to form an initial population; performing genetic operations on parent populations in the initial population by a hovering energy-saving strategy to generate a child population according to the fitness function; merging the child population and the parent populations to form a merged population; calculating the fitness of all individuals in the merged population and sorting the individuals in descending order according to the fitness; selecting a preset number of optimal individuals based on the result of the descending order to form a target population; outputting the target population as an optimal unmanned aerial vehicle energy consumption trajectory strategy until the fitness converges or a preset maximum number of generations is reached; and executing the optimal unmanned aerial vehicle energy consumption trajectory strategy.

The application discloses a wheat scab prediction method based on multi-source data and an ES-GBM algorithm, and comprises the following steps: S1, collecting field wheat scab disease grade sample point data of a target area; S2, acquiring multi-source remote sensing data of the target area, wherein the multi-source remote sensing data at least include multiple characteristic variables in atmospheric parameters, weather parameters, vegetation spectrum parameters and atmospheric environment parameters; S3, using the sample point data and sample point attribute data extracted from the multi-source remote sensing data, constructing a prediction model based on an evolutionary strategy ES optimizing a gradient boosting machine GBM, wherein the ES is used for automatically searching an optimal hyperparameter combination of the GBM; and S4, inputting gridded multi-source remote sensing data of the target area into the prediction model to obtain a wheat scab disease degree spatial distribution map of the target area. The application can fuse multi-source data, and through the evolutionary strategy, an optimal hyperparameter combination is searched, so that the prediction precision of the model is improved.

The application relates to an energy-driven crowd optimization method and device, electronic equipment and a readable storage medium, which comprises the following steps: generating an initial population corresponding to a robot obstacle avoidance path, initializing each individual in the initial population; determining the individual fitness of each individual, determining an elite set and an exploration set, and determining the dynamic role parameters of each individual; determining the first target orientation of the individual in the elite set and the second target orientation of the individual in the exploration set, thereby adjusting the development direction and the exploration direction corresponding to the elite set and the exploration set; selecting a differential evolution strategy or an energy-driven behavior strategy to update the individual position with a preset probability, obtaining an updated individual position; updating the energy value corresponding to the individual based on the updated individual position, determining an optimal population corresponding to a global optimal solution, and obtaining a robot obstacle avoidance path. Through the method provided by the application, the global exploration and local development capabilities are enhanced, and the adaptability and search efficiency in a dynamic environment are improved.

The present application relates to the technical field of workshop scheduling, in particular to a multi-modal multi-objective reentrant hybrid flow shop scheduling method, which is based on a multi-modal multi-objective meme method driven by hierarchical online learning, and comprises the following steps: step 1: establishing a problem model; step 2: setting method running parameters; step 3: generating a population by using an initialization strategy; step 4: judging whether a first-stage termination condition is met, if not, executing a first-stage evolution strategy and an environment selection strategy on the population, otherwise executing step 5; step 5: constructing a multi-modal archive; step 6: judging whether a second-stage termination condition is met; the present application can effectively mine a large number of performance-equivalent but structurally different multi-modal scheduling solutions while guaranteeing the convergence of the target space and maintaining the diversity of the decision space.

The application discloses a shot key frame segmentation and fusion method based on multi-level cooperation and intelligent node management, and relates to the technical field of video data processing. The method comprises the following steps: S100, extracting an original image frame sequence, using a feature extraction operator to extract color / gradient feature vectors of each frame of image, and calculating the similarity distance between any two frames, and identifying a video semantic boundary through an improved dynamic time warping algorithm, so that the original video stream is divided into multiple video segments; S200, constructing a multi-level perception architecture based on the video segments, and respectively deploying a basic perception layer, a time sequence correlation layer and a global semantic layer on edge nodes corresponding to an algorithm power gradient, synchronizing intermediate results through a message queue and cooperatively training, so that the accurate calculation of a video shot boundary probability is realized. The application realizes the deep fusion of high precision and processing timeliness of video semantic boundary capture by constructing a heterogeneous node cluster at a physical level and implementing hierarchical perception and federal evolution strategies at a logical level.