Product
Eureka
For R&D, Eureka makes reading and utilizing patents & technical documents easy.
Eureka AIR
Designed for self-driven R&D workflows. Generate viable solutions, solve complex R&D challenges, empower your innovation with AI.
Eureka Materials
Designed for material experts only. Revolutionize your material R&D, from search, analyze, to developing new materials.
Feature
TechResearch
Generate reliable direction feasibility study reports for your R&D in just a few steps.
TechSeek
Discover and master advanced knowledge NOW. Basics, ideas, possibilities, all at once.
TechMind
As an expert in R&D Theories, TechMind can generates customized viable solutions instantly.
TechRisk
Analyze your overall solution with one click, know your potential R&D risks in advance.
TechMonitor
Get weekly tech updates, stay abreast of the latest tech innovations and key insights.
Patents
Literature
Hiro is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
59 results about "Coevolution" patented technology
Efficacy Topic
Property
Owner
Technical Advancement
Application Domain
Technology Topic
Technology Field Word
Patent Country/Region
Patent Type
Patent Status
Application Year
Inventor
In biology, coevolution occurs when two or more species reciprocally affect each other's evolution. Charles Darwin mentioned evolutionary interactions between flowering plants and insects in On the Origin of Species (1859). The term coevolution was coined by Paul R. Ehrlich and Peter H. Raven in 1964. The theoretical underpinnings of coevolution are now well-developed, and demonstrate that coevolution can play an important role in driving major evolutionary transitions such as the evolution of sexual reproduction or shifts in ploidy. More recently, it has also been demonstrated that coevolution influences the structure and function of ecological communities as well as the dynamics of infectious disease.
Flexible job shop scheduling method based on multi-species coevolution
InactiveCN101901425AQuality improvementAvoid the disadvantages that the performance cannot be fully utilizedResourcesGuidelineJob shop scheduling
The invention provides a flexible job shop scheduling method based on multi-species coevolution, which belongs to the field of shop scheduling, and mainly overcomes the disadvantage that a flexible job shop scheduling method based on a genetic algorithm can not be exerted fully. The method comprises the main steps of: 1. setting parameters; 2. initializing species according to a set coding method; 3. calculating the fitness value of each chromosome in each species according to a set method, and recording the optimum fitness value and the constitute chromosomes thereof; 4. carrying out multi-species coevolution: carrying out evolution operations, i.e. chiasmata and variation, to chromosomes in each subspecies, and evaluating new chromosomes; and 5. judging whether a method termination criteria is achieved or not: if so, terminating the method and outputting the optimum fitness value and the constitute chromosomes thereof; and otherwise, jumping to the step 4. The invention can be used to obtain a high-quality scheduling scheme suitable for practical production of shops, can shorten production time, and can be used for scheduling management and optimization of the production process of shops.
Owner:HUAZHONG UNIV OF SCI & TECH
Robot path planning method based on coevolution particle swarm rolling optimization
InactiveCN103336526ARealize autonomous navigationImprove search capabilitiesVehicle position/course/altitude controlPosition/direction controlEngineeringControl theory
The invention discloses a robot path planning method adopting multi-subgroup coevolution particle swarm optimization to perform real-time rolling optimization obstacle avoidance processing, which can enable a robot to realize autonomous navigation in a working condition with stationary obstacles at unknown positions. The method has the following advantages: the method can realize autonomous navigation of the robot in a working condition with stationary obstacle at unknown position; the method introduces group mass center strategy to the multi-subgroup coevolution, so as to increase the search capacity of the population; compared with a conventional path planning method which adopts PSO (Particle Swarm Optimization) and CPSO (Coevolution Particle Swarm Optimization), the method provided by the invention adopts survival of the fittest strategy, thereby being more advantageous in respect to rate and accuracy of evolution convergence; the method provided by the invention can real-timely plan an effective obstacle avoidance path in a complex obstacle environment, and has the advantages of good robustness, high solving efficiency, and the like.
Owner:SUZHOU INST OF TRADE & COMMERCE
Optimization design method of radial-flow-type hydraulic turbine
ActiveCN102608914ASmall amount of calculationHigh precisionAdaptive controlImpellerMultivariable optimization
Owner:开山(西安)透平机械有限公司
Reactive optimizing method of power system based on coordinate evolution
InactiveCN1564416ATo achieve the purpose of optimizationReactive power adjustment/elimination/compensationReactive power compensationElectric power systemOriginal data
The method includes following steps: dividing control variables (as plant or animal communities) for system reactive optimization into several groups, and each variable is corresponding to a flock in coevolution method; inputting original data, initializing each flock, calculating adaptation function value of each chromosome in initial flocks; based on previous generation, generating new generation through genetics operations of selection, chiasma, variation; calculating adaptation function value fro new generation; selecting optimum chromosome; evolutionary optimization of ecosystem is completed after optimizing each flock; determining whether condition of convergence of genetic algorithm is met, and outputting optimized result. In the invention, problem to be solved is mapped to ecosystem including multiple flocks. Coevolution of interactive flocks makes optimization of system.
Owner:XI AN JIAOTONG UNIV
Automobile part welding optimal path planning method of multi-robot coordination
ActiveCN107390684AIncrease diversityImprove global search performancePosition/course control in two dimensionsGeneration processGenetic algorithm
The invention discloses an automobile part welding optimal path planning method of multi-robot coordination. The method comprises a welding spot information preprocessing step, a welding spot path planning step of a hybrid genetic-particle swarm optimization algorithm based on coevolution and a motion path step in which welding spot path planning information is converted into a welding robot. In the method, a genetic algorithm and a particle swarm optimization algorithm are combined; based on a mutation characteristic of the genetic algorithm, a diversity of a particle swarm during a particle swarm algorithm iteration process and a global search capability can be increased; and a welding path optimal planning problem during an automobile part generation process can be solved.
Owner:SOUTH CHINA UNIV OF TECH
Method for generating test data covering parallel program paths based on coevolution
InactiveCN103902457AGenerate efficientlyGenetic modelsSoftware testing/debuggingMathematical modelTheoretical computer science
The invention provides a method for generating test data covering parallel program paths based on coevolution, and aims to provide a method for automatically and efficiently generating test data covering parallel program objective paths. The method includes the following specific steps that firstly, a mathematical model of a test data generation problem is built, and a problem for generating the test data covering the parallel program paths is modeled into a single-object optimization problem; secondly, a coevolution genetic algorithm is designed to solve the model. According to the method, groups are divided into a plurality of sub groups and a cooperative team group according to the correlation of course paths and program input components. Each sub group is used for independently optimizing a part of input components relevant to one certain course path. After the sub groups are evolved into a certain period, excellent individuals of the sub groups are combined to form an initial individual of the cooperative team group so as to be used for optimizing complete program input. After the cooperative group is evolved into a certain period, the excellent individuals are returned to the sub groups. Through alternate coevolution of the cooperative team group and the sub groups, the expected test data are generated.
Owner:CHINA UNIV OF MINING & TECH
Social network-oriented multi-information and multi-dimensional network information propagation model and method
ActiveCN108230170AIn line with the real situationData processing applicationsWeb data retrievalInformation DisseminationMulti dimensional
The invention discloses a social network-oriented multi-information and multi-dimensional network information propagation model and method, and belongs to the field of social network analysis. The method comprises the following steps of: firstly, obtaining social network data and preprocessing the data; secondly, extracting user information, user behaviors and user relationships from real data, and constructing a multi-dimensional network space by using a cosine similarity method; thirdly, establishing a model, importing influence factors on the basis of traditional epidemic models through using an epidemic model mechanism for reference, so as to express interaction relationships and intensities between different pieces of information, and then constructing a multi-information and multi-dimensional space network-based information propagation model; and finally, carrying out simulation analysis, constructing a kinetic equation from a micro perspective and a macro perspective so as to analyze a common evolution trend of two messages. The model and method more accord with real scenes of information propagation and are more beneficial for research of information propagation processes.
Owner:CHONGQING UNIV OF POSTS & TELECOMM
Multi-target distributed power source site selection constant volume method considering power supply reliability
InactiveCN108446805AImprove scienceOvercoming simple weighting for multiple objectivesForecastingPower BalanceDistributed power
The present invention provides a multi-target distributed power source site selection constant volume method considering power supply reliability. The method comprises the steps of: performing calculation of a plurality of target functions of a distributed power source site selection constant volume, integrating the target functions to establish a target function of the distributed power source site selection constant volume considering the power supply reliability, combining a distribution network power balance constraint condition, a distributed power source output constraint condition and adistribution network node voltage constraint condition to establish an optimization model of the multi-target distributed power source site selection constant volume considering the power supply reliability, employing the particle swarm optimization and the non-dominated sorting coevolution algorithm to perform model solution, and obtaining a distributed power source site selection constant volume optimization scheme of multiple targets. The defects are overcome that a current site selection constant volume algorithm is not practical to perform multi-target and simple weighing, the multi-target distributed power source site selection constant volume method is simple in calculation and fast in rate of convergence, improves the scientificity of the distributed power source site selection constant volume scheme, relieves the influence of the distributed power source for the distribution network operation and improves the economical efficiency of the distributed power source.
Owner:STATE GRID JIANGXI ELECTRIC POWER CO LTD RES INST +2
Method for domestication of salt-resistant activated sludge
ActiveCN103910475AEasy to handleImprove settlement performanceSludge processingBiological sludge treatmentActivated sludgeSalinity
The invention provides a method for domestication of salt-resistant activated sludge. The method comprises that substances (such as soil or water) in a high-salt environment are directly added into activated sludge having a normal salt concentration according to a certain ratio and then salt concentration loading is gradually increased, and in the above process, the microflora in the high-salt environment and the microflora in the activated sludge carry out bidirectional coevolution so that novel microflora is formed. The method is free of a complex bacterium screening process, has simple and easy processes, utilizes a novel salinity gradient-increasing scheme, prevents a biomass loss caused by overquick increasing of salinity pressure and realizes fast and efficient domestication of high-salt sludge.
Owner:XINJIANG ENVIRONMENTAL ENG TECH
Integrated optimization method for arch truss chip mounter based on coevolution
The invention provides an integrated optimization method for an arch truss chip mounter based on coevolution. The method mainly comprises the following steps: (1) building a suction nozzle configuration optimization model and an integrated optimization mathematical model combining feeder allocation and component bonding order; (2) utilizing linear program to solve the suction nozzle configuration optimization model; (3) conducting coevolution on the feeder allocation and the component patching order adopting evolution strategies of neighborhood competition, cross-connection, variation and partial search based on coevolution to enable moving path of a bonding head to be minimum during a bonding process. The integrated optimization method has the advantages that working time of the chip mounter is shortened, and bonding efficiency is effectively improved; the method can be applied to optimization control of the arch truss chip mounter during a surface assemble process; the method overcomes the defect that conventional optimization methods are instable and simplex to solve a complex multi-decision optimization problem, and adopts coevolution to conduct simultaneous optimization on multiple subproblems.
Owner:SOUTH CHINA UNIV OF TECH
In-vitro directed coevolution method for modifying L-phenylalanine gene engineering strains
InactiveCN101698853ASpeed up evolutionHigh catalytic activityMicroorganism based processesFermentationPhenylalanineMetabolic balance
The invention relates to an in-vitro directed coevolution method for modifying L-phenylalanine gene engineering strains, which is realized in a way that: using genes aroG and pheA on an L-phenylalanine gene engineering strain constructed in the room as a whole; carrying out the in-vitro directed coevolution modification by using an error-prone PCR technology and a recombinant DNA technology; and screening to obtain a mutant strain, of which the yield of L-phenylalanine is increased by 114%. In the invention, by modifying the gene formed by coupling and connecting aroG and pheA in series, the key genes aroG and pheA in the metaboly process of the L-phenylalanine are used as a whole to carry out the directed modification, thereby obtaining a new metabolic balance, so that the expressed enzyme has high-efficiency catalytic activity and can resist the feedback inhibition of the L-phenylalanine, thereby obtaining the new modified high-yield L-phenylalanine gene engineering strain by screening. The method can provide an example for modifying the acid production rate of any amino acid gene engineering strain.
Owner:MAIDAN BIOLOGICAL GROUP FUJIAN
Automated scalable and adaptive system for memory analysis via the discovery of co-evolving regions
A method for identifying co-evolving regions in the memory of a target application comprises the steps of: receiving information identifying a set of data structures that are evolving; and classifying the constituents of the data structures based on their likelihood to evolve in a single coherent manner. Classifying the constituents of the data structures based on their likelihood to evolve in a single coherent manner further comprises determining coherency by similarity of key structural features and data type of the constituents. The method also comprises qualitative and quantitative characterizations of the co-evolving regions based on their observed evolutions.
Owner:IBM CORP
Application method of rastonia solanacearum bacteriophage
ActiveCN104542717AReduce the amount of applicationPromote absorptionBiocideDisinfectantsDiseaseVascular bundle
The invention discloses an application method of rastonia solanacearum bacteriophage. The application method comprises the following steps: placing rastonia solanacearum bacteriophage into a sterile syringe needle diagonally inserted in a tobacco plant stem, then covering the rastonia solanacearum bacteriophage with sterile mineral oil so as to prevent evaporation and pollution, so that the rastonia solanacearum bacteriophage directly enters the tobacco plant stem through the sterile syringe needle. According to the characteristic that the rastonia solanacearum is typical bacterial vascular bundle disease, the application method for utilizing the bacteriophage resources to prevent rastonia solanacearum disease is designed, and the method is applied to prevention and control of bacterial wilt of crops, the application amount of the rastonia solanacearum bacteriophage can be reduced, the influence of the outside factors such as sunshine, ultraviolet rays, and rainwater can be avoided, and the resistance on the bacteriophage obtained through the coevolution of the pathogen rastonia solanacearum and the rastonia solanacearum bacteriophage due to long-time contact can be avoided.
Owner:GUIZHOU TOBACCO SCI RES INST
Bicycle-mode traveling selection forecasting method based on activity chain mode
InactiveCN103116702AScientific guidanceReasonable guidanceSpecial data processing applicationsTravel modeSimulation
The invention discloses a bicycle-mode traveling selection forecasting method based on an activity chain mode. The forecasting method includes the steps that the data survey is carried on a situation of resident traveling, and the survey result is managed and added up; a selecting mode of the resident traveling in a day in the data survey result is extracted, and the traveling mode is carried out on a variable virtual operation and a coding operation; a correlated variable in the activity chain mode is input to multi-term logit models, and a coevolution logit model can be obtained through calculating; the calculated coevolution logit model is carried out on iterative operation, and two selecting results of traveling modes are recorded; the two selecting results of the traveling modes are carried out on statistics and analysis, the prediction accuracy is carried out on contrastive analysis. By the statistics and the analysis of the vehicle selection of residents in urban, the proportion of the bicycle-mode traveling selection can be accurate to forecast, so that the urban traffic planning and the decision of the policy can be provided with the scientific and reasonable guidance.
Owner:SOUTHEAST UNIV
Multi-population coevolution method for optimizing wireless sensor network topology
ActiveCN106789320ANo change in scale-free propertiesThe initial degree does not changeNetwork topologiesData switching networksAlgorithmNetwork topology
The invention relates to the field of robustness optimization of network topology and provides a multi-population coevolution method for optimizing wireless sensor network topology. The method comprises the following steps: (S100) generating initial individuals of each population based on source topology; (S200) selecting parent individuals from each population, and carrying out crossover operator operation, so as to generate new filial generations; (S300) randomly selecting a certain number of individuals from the individuals of the populations, and carrying out mutation operator operation; (S400) after crossover mutation of each population is finished, carrying out fitness function screening, and selecting outstanding individuals for the next generation; (S500) carrying out immigration among the populations, so as to generate gene exchange among the populations; and (S600) circularly carrying out the steps (200) to (500), storing the most individual topology in each generation, judging the number of evolution generations, and exiting the circulation after reaching a set number of evolution generations.
Owner:DALIAN UNIV OF TECH
Multi-path coverage test data coevolution generation method for message-passing parallel program
ActiveCN108549607AImprove production efficiencyImprove automationSoftware testing/debuggingGenetic algorithmsGenetic algorithmMessage passing
The invention discloses a multi-path coverage test data coevolution generation method for a message-passing parallel program, and aims to efficiently generate test data covering multiple target pathsfor the message-passing parallel program. The method comprises the following specific steps of (1) constructing corresponding populations for the target paths in each scheduling sequence respectively,wherein individuals in the populations are coded program inputs; (2) designing population performance and individual performance evaluation indexes; (3) solving the populations by using a genetic algorithm, and migrating the individuals to the populations with good performance through individual migration in the process; and (4) stopping evolution of the populations corresponding to the covered target paths according to evolution results of all generations, and stopping an algorithm until the test data covering all the target paths is generated or maximum evolution generations are reached.
Owner:CHINA UNIV OF MINING & TECH
Genome unit point variation pathogenicity prediction method and system and storage medium
ActiveCN110245685APrivacy protectionLow costBiostatisticsCharacter and pattern recognitionFeature extractionAlgorithm
The invention relates to the technical field of bioinformatics, and provides a genome unit point variation pathogenicity prediction method and system and a storage medium. The method comprises the following steps: obtaining genome unit point variation data and coevolution conservative data according to a genome unit point variation position and a variation condition; preprocessing the genome unit point variation data and the coevolution conservative data to generate a matrix; loading the model, inputting the matrix, performing feature extraction through the densely connected convolutional neural network, splicing the feature data by adopting the multilayer perceptron, performing calculation, and outputting a prediction result. By adopting the method, the problems of low prediction accuracy, low reliability and high cost of genomic unit point variation pathogenicity in the prior art can be solved.
Owner:TSINGHUA UNIV
Specific primer pair for assisting identification of Klebsiella pneumoniae
InactiveCN102367483AImprove detection efficiencyHigh sensitivityMicrobiological testing/measurementMicroorganism based processesK pneumoniaeAgricultural science
The invention discloses a specific primer pair for assisting identification of Klebsiella pneumoniae. The primer pair provided in the invention consists of DNA shown in sequence 1 of the sequence table and DNA shown in sequence 2 of the sequence table. The specific primer pair designed in the invention can be used for assisting identification about whether Bactrocera dorsalis (Hendel) carries Klebsiella pneumoniae only by means of PCR (polymerase chain reaction) and electrophoresis, and the identification is characterized by high sensitivity, simple process and no need for cloning. In order to increase accuracy, a PCR amplified product can be subjected to sequencing. The specific primer pair provided in the invention improves the efficiency of Bactrocera dorsalis (Hendel) symbiotic bacteria (Klebsiella pneumoniae), and creates conditions for further study of the coevolution relation between Bactrocera dorsalis (Hendel) and Klebsiella pneumoniae.
Owner:CHINA AGRI UNIV
Differential-evolution protein-structure head-beginning prediction method based on multistage sub-population coevolution strategy
ActiveCN106503486AIncrease diversityHigh precisionProteomicsGenomicsPredictive methodsSub populations
The invention discloses a differential-evolution protein-structure head-beginning prediction method based on the multistage subpopulation coevolution strategy. The differential-evolution protein-structure head-beginning prediction method includes the following steps that under a differential-evolution algorithm framework, the conformational space dimensionality is reduced through a Rosetta Score3 coarse-granularity knowledge energy model; an evolution population is divided into a plurality of subpopulations according to the similarity, coevolution is carried out on the subpopulations, and the individual diversity of the population can be improved; the evolutionary process is divided into three stages, different variation crossover strategies are adopted at different stages, and the premature convergence problem can be solved; the conformational space can be effectively sampled in cooperation with the high global searching ability of the differential-evolution algorithm, and the high-accuracy conformation close to the natural state is obtained through searching. Based on the differential-evolution algorithm, the differential-evolution protein-structure head-beginning prediction method based on the multistage subpopulation coevolution strategy is low in conformational space searching dimension and high in convergence speed and prediction accuracy.
Owner:ZHEJIANG UNIV OF TECH
Asynchronous Evaluation Strategy For Evolution Of Deep Neural Networks
The technology disclosed proposes a novel asynchronous evaluation strategy (AES) that increases throughput of evolutionary algorithms by continuously maintaining a queue of K individuals ready to be sent to the worker nodes for evaluation and evolving the next generation once a fraction Mi of the K individuals have been evaluated by the worker nodes, where Mi<<K. A suitable value for Mi is determined experimentally, balancing diversity and efficiency. The technology disclosed is extended to coevolution of deep neural network supermodules and blueprints in the form of AES for cooperative evolution of deep neural networks (CoDeepNEAT-AES). Applied to image captioning domain, a threefold speedup is observed on 200 graphics processing unit (GPU) worker nodes, demonstrating that the disclosed AES and CoDeepNEAT-AES are promising techniques for evolving complex systems with long and variable evaluation times.
Owner:COGNIZANT TECH SOLUTIONS U S CORP
Adaptive coevolution algorithm-based information kernel extraction method
InactiveCN107609033ACheck the recommended results are excellentThe recommendation effect is accurateGenetic modelsSpecial data processing applicationsMutation operatorAlgorithm
The invention discloses an adaptive coevolution algorithm-based information kernel extraction method, and mainly solves the problem of incapability of ensuring an optimal recommendation effect in theprior art. The method is implemented by the steps of (1) constructing a user score matrix; (2) initializing parent populations; (3) adaptively adjusting a cross operator selection probability; (4) adaptively adjusting a mutation operator selection probability; (5) classifying the parent populations; (6) building a team; (7) judging whether each team member comes from a parent elite population or not; (8) updating progeny elite populations; (9) updating progeny ordinary populations; (10) judging whether all information kernels of the parent elite population are central information kernels or not; (11) updating the parent populations; (12) judging whether an iterative frequency is equal to 200 or not; and (13) outputting an optimal information kernel. An experimental simulation result showsthat compared with an existing information kernel extraction method, the method provided by the invention has a better recommendation effect.
Owner:XIDIAN UNIV
Method for enhancing leadership, entrepreneurship, performance, innovation, creativity, and career achievement.
A method for cultivating a dynamic foundation of leader drives, reflexes, and meta-competencies through a paradigm shift to a paradigm composed of a logically integrated system of dynamics. These dynamics pertain to how systems synergistically adapt, advance and co-evolve to attain and sustain peak performance. The paradigm shift is achieved through a series of presentations, quantum leaps, belief upgrades, exercises, and action-learning experimentation which provide component systems of beliefs and information until the full paradigm has been assimilated. The invention launches a series of self-motivating growth continuums which capitalize on natural mechanisms and drives to trigger life-long development. The dynamic foundation improves the assimilation and use of classic leader competencies taught by traditional leadership development and performance improvement programs to create more impactful leaders, entrepreneurs, innovators, and individuals.
Owner:HOLMES LAUREN L
Methods and Systems for Identification of Biomolecule Sequence Coevolution and Applications Thereof
Generation of biomolecule sequence coevolution data structures, matrices, scores, and sectors are described. Generally, the generated coevolution data removes covariant noise due to phylogenetic drift and can reveal coevolution of residue positions in multiple phylogenetic distances. Scores can be built upon the data structures and matrices to reveal sectors of residue positions that function and evolve together. Furthermore, the coevolution data structures, matrices, scores, and sectors can be used to predict structure or function of residue variants.
Owner:THE BOARD OF TRUSTEES OF THE LELAND STANFORD JUNIOR UNIV
Network community detection method based on M elite coevolution strategy
InactiveCN103457800ADiversity guaranteedAccurate detectionData switching networksCommunity structureStructure of Management Information
The invention discloses a network community detection method based on an M elite coevolution strategy, wherein the network community detection method solves the problems that in the prior art, the convergence rate is low and easily lapses into the local optimum, multiresolution analysis of a network structure cannot be achieved. The implementation steps include that (1) network data are loaded; (2) network community populations are initialized; (3) the network community populations are divided; (4) a network community team is organized; (5) candidate network community division is detected; (6) the network community populations are updated; (7) local network communities are detected; (8) the network community populations are updated; (9) whether iteration is terminated or not is judged; (10) a network community detection result is output. When the network community detection method is used for detecting community structures in a network, expanded module density functions serve as fitness functions, a network structure is analyzed with different resolutions, and the convergence rate is quickened through leading-in of local detection and does not easily lapse into the local optimum.
Owner:XIDIAN UNIV
Multi-recycle-station garbage collection and transportation method based on coevolution
ActiveCN110516870AImprove efficiencyOvercoming the early convergence problemForecastingCharacter and pattern recognitionCluster algorithmRefuse collection
The invention provides a multi-recycle-station garbage collection and transportation method based on coevolution, and the method comprises the steps: employing a CC-HGA to improve a clustering algorithm, distributing each garbage collection point to a proper recycle station, and converting MSRCP into a plurality of single-recycle-station garbage collection and transportation problems; the method specifically comprises eight steps to realize garbage collection and transportation at multiple recycle stations. The method has the advantages that the improved clustering algorithm is combined with the CC framework to decompose the solution space, better grouping is provided, evolution of the whole population is guided through evolution cooperation of the sub-populations, the efficiency of the algorithm is effectively improved, and the high-dimensional problem solving capacity of the algorithm is improved; according to the hybrid genetic algorithm, a local search operator is improved, the understanding search range is expanded, and the early convergence problem encountered when the problem is solved by the algorithm is solved.
Owner:ANQING NORMAL UNIV
Deep neural network optimizing method based on coevolution and back propagation
InactiveCN106650933AOptimization parametersEfficient training completedNeural learning methodsLocal optimumAlgorithm
The invention discloses a deep neural network optimizing method based on coevolution and back propagation, belongs to the technical field of combination of deep learning and evolutionary computation, and mainly aims to solve the problem of tendency to fall into a local optimal solution during training of the deep neural network. The deep neural network optimizing method comprises the following steps: (1) firstly, optimizing a network by using a back propagation algorithm; (2) when a stopping condition is satisfied, optimizing by using a co-evolution algorithm; (3) continually repeating the previous steps until an iterative stopping condition is satisfied; and (4) performing final optimization to obtain a weight and a deviation as optical parameters. In the deep neural network optimizing method, the advantages of an evolution algorithm are applied to the training of the deep neural network, and large-scale parameters are optimized by coevolution. Meanwhile, a selection strategy is designed in conjunction with the back propagation algorithm in order to increase the optimizing speed of the coevolution, so that the whole network can be trained more efficiently, high optimizing performance is achieved, and the classification accuracy of the network is increased.
Owner:XIDIAN UNIV
Multi-species coevolution method for solving warehousing operation optimization problem with aisles
ActiveCN110033121AIncrease diversityTaking into account the breadthForecastingArtificial lifeLogistics managementPredation
The invention discloses a multi-species coevolution method for solving the storage operation optimization problem with aisles, and belongs to the field of intelligent logistics and storage equipment.According to the method, the transverse aisles are added on the basis of traditional warehousing, the warehousing operation optimization model with the aisles is established, and the model is solved.In view of the defects that the existing solving technology is easy to premature and low in convergence speed, the invention provides a multi-species co-evolution optimization method based on the joint participation of a genetic algorithm, a particle swarm algorithm and an artificial fish swarm algorithm, i.e., through a multi-species competition symbiotic predation strategy based on a learning mechanism, the environment adaptability of each species can be enhanced; by introducing a variation mechanism, the population diversity of all species is synergistically improved, so that the evolutioncapability of a single species is improved, and meanwhile, the global optimization capability and the solving efficiency of the algorithm are also improved. According to the method, the operation efficiency of overall storage is improved, and the logistics storage can be promoted to be transformed and upgraded to be intelligent and green.
Owner:HENAN INST OF SCI & TECH
Multi-objective optimization method based on double-layer elite coevolution
PendingCN111046559AAddressing the Insufficient Non-Dominated Individual SituationFast convergenceArtificial lifeDesign optimisation/simulationAlgorithmTheoretical computer science
The invention discloses a multi-objective optimization method based on double-layer elite coevolution, and solves the problems of non-uniform distribution of elite individuals at the initial stage andlow convergence rate in the solving process of a multi-objective problem. According to the method, a two-layer elite population division strategy is adopted to solve the problem of non-uniform distribution of excellent individuals in the initial stage; by adopting the coevolution method, the cooperation capability among individuals can be fully exerted, and the diversity and convergence of the individuals in the evolution process are guaranteed; a probability model is established by adopting a distribution estimator, the evolution trend of the whole group is directly described, and the globalsearch capability of the method can be guaranteed.
Owner:南京邮电大学通达学院
Characteristic optimization method based on coevolution for foot passenger detection
InactiveCN101246555AReduce computational complexitySolve problems that are hard to decomposeCharacter and pattern recognitionAlgorithmSimulation
The invention relates to a feature optimization selection method for a pedestrian detection based on coevolution, which includes that: (1) a training sample is read in; (2) an original characteristic set is generated and a sample set is formed; (3) four populations are initialized and a type of characteristic is corresponded to each population; (4)an individual is decoded to a feature combination and then a new sample subset is obtained, and fitness of the individual is calculated; (5) a terminal condition is judged for whether the requirement is met, if the terminal condition is met, a characteristic subset denoted by a best individual in each population is used as the optimum relation of an algorithm; (6) a competition within the population, an inter-population competition and self-increase rules are used for choosing the individual according to the fitness of each individual, a method for single interior extrapolation and its variation are used for generating the next generation individual; (7) the (4) step is returned and the population is evolved until an feature selection terminal condition of the (5) step is satisfied. The invention decreases the complexity of computation, and can obtain an optimizing feature subset, and promotes the veracity for pedestrian classification.
Owner:UNIV OF SCI & TECH OF CHINA
Ontology matching method based on compact coevolution algorithm
InactiveCN106227798AOvercoming the defects of premature convergenceQuality improvementSpecial data processing applicationsInternal memoryAlgorithm
An ontology matching method based on a compact coevolution algorithm comprises the steps that an optimization model of an ontology matching problem is established, and a similarity matrix is established; the optimization model is solved by the compact coevolution algorithm, and an optimal ontology matching result is acquired; probability vectors PV_better of better individuals and probability vectors PV_worse of worse individuals are initialized; an elite solution ind_BElite of the initialized probability vectors of the better individuals, as well an elite solution ind_Bworse of the initialized probability vectors of the worse individuals are generated according to the PV_better and PV_worse; coding information of each individual comprises a weight used to integrate mapping results of different similarity measures as well as a threshold value used to filter ontology mapping results; and a weighted averaging method is used to integrate matching results generated from the different similarity measures. According to the invention, time and the internal memory amount consumed during running of an ontology matching system based on an evolution algorithm can be essentially reduced, so that efficiency of ontology matching can be increased.
Owner:FUZHOU INSTITUE OF TECH
Who we serve
- R&D Engineer
- R&D Manager
- IP Professional
Why Eureka
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Social media
Eureka Blog
Learn More Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com