A Genetic Algorithm-Ant Colony Algorithm(GA-ACA),which can be used to optimize multi-Unit Under Test(UUT)parallel test tasks sequences and resources configuration quickly and accurately,is proposed in the paper.With t...A Genetic Algorithm-Ant Colony Algorithm(GA-ACA),which can be used to optimize multi-Unit Under Test(UUT)parallel test tasks sequences and resources configuration quickly and accurately,is proposed in the paper.With the establishment of the mathematic model of multi-UUT parallel test tasks and resources,the condition of multi-UUT resources mergence is analyzed to obtain minimum resource requirement under minimum test time.The definition of cost efficiency is put forward,followed by the design of gene coding and path selection project,which can satisfy multi-UUT parallel test tasks scheduling.At the threshold of the algorithm,GA is adopted to provide initial pheromone for ACA,and then dual-convergence pheromone feedback mode is applied in ACA to avoid local optimization and parameters dependence.The practical application proves that the algorithm has a remarkable effect on solving the problems of multi-UUT parallel test tasks scheduling and resources configuration.展开更多
Rainfall erosivity in Tibet from 2000 to 2OlO was estimated based on simplified erosion prediction model using daily rainfall data derived from the Tropical Rainfall Measurement Misssion （TRMM） 3B42 rainfall measure...Rainfall erosivity in Tibet from 2000 to 2OlO was estimated based on simplified erosion prediction model using daily rainfall data derived from the Tropical Rainfall Measurement Misssion （TRMM） 3B42 rainfall measurement algorithm. Semi- monthly erosive rainfall and rainfall erosivity were validated using weather station data. The spatial distribution of annual rainfall erosivity as well as its seasonal and annual variation in Tibet was also examined. Results showed that TRMM 3B42 data could serve as an alternative data source to estimate rainfall erosivity in the area where only data from sparsely distributed weather stations are available. The spatial distribution of rainfall erosivity in Tibet generally resembles the distribution of multi-year average of annual rainfall. Annual rainfall erosivity in Tibet decreased from the southeast to the northwest. The concentration degree of rainfall erosivity shows an increasing trend from the southeast to the northwest. High rainfall erosivity accompanies low rainfall erosivity concentration degree and vice versa. Rainfall erosivity increased in the middle and western Tibet and decreased in the southeastern Tibet during the 11 years of this study.展开更多
Deep neural networks have been successfully applied to numerous machine learning tasks because of their impressive feature abstraction capabilities.However,conventional deep networks assume that the training and test ...Deep neural networks have been successfully applied to numerous machine learning tasks because of their impressive feature abstraction capabilities.However,conventional deep networks assume that the training and test data are sampled from the same distribution,and this assumption is often violated in real-world scenarios.To address the domain shift or data bias problems,we introduce layer-wise domain correction(LDC),a new unsupervised domain adaptation algorithm which adapts an existing deep network through additive correction layers spaced throughout the network.Through the additive layers,the representations of source and target domains can be perfectly aligned.The corrections that are trained via maximum mean discrepancy,adapt to the target domain while increasing the representational capacity of the network.LDC requires no target labels,achieves state-of-the-art performance across several adaptation benchmarks,and requires significantly less training time than existing adaptation methods.展开更多
基金supported by“11th Five-year Projects”pre-research projects fund of the National Arming Department
文摘A Genetic Algorithm-Ant Colony Algorithm(GA-ACA),which can be used to optimize multi-Unit Under Test(UUT)parallel test tasks sequences and resources configuration quickly and accurately,is proposed in the paper.With the establishment of the mathematic model of multi-UUT parallel test tasks and resources,the condition of multi-UUT resources mergence is analyzed to obtain minimum resource requirement under minimum test time.The definition of cost efficiency is put forward,followed by the design of gene coding and path selection project,which can satisfy multi-UUT parallel test tasks scheduling.At the threshold of the algorithm,GA is adopted to provide initial pheromone for ACA,and then dual-convergence pheromone feedback mode is applied in ACA to avoid local optimization and parameters dependence.The practical application proves that the algorithm has a remarkable effect on solving the problems of multi-UUT parallel test tasks scheduling and resources configuration.
基金supported by the Natural Science Foundation of China (Grant No. 40925002)the National Science and Technology Supporting Program in the Eleventh Five-Year Plan of China (Grant No. 2007BAC06B06)
文摘Rainfall erosivity in Tibet from 2000 to 2OlO was estimated based on simplified erosion prediction model using daily rainfall data derived from the Tropical Rainfall Measurement Misssion （TRMM） 3B42 rainfall measurement algorithm. Semi- monthly erosive rainfall and rainfall erosivity were validated using weather station data. The spatial distribution of annual rainfall erosivity as well as its seasonal and annual variation in Tibet was also examined. Results showed that TRMM 3B42 data could serve as an alternative data source to estimate rainfall erosivity in the area where only data from sparsely distributed weather stations are available. The spatial distribution of rainfall erosivity in Tibet generally resembles the distribution of multi-year average of annual rainfall. Annual rainfall erosivity in Tibet decreased from the southeast to the northwest. The concentration degree of rainfall erosivity shows an increasing trend from the southeast to the northwest. High rainfall erosivity accompanies low rainfall erosivity concentration degree and vice versa. Rainfall erosivity increased in the middle and western Tibet and decreased in the southeastern Tibet during the 11 years of this study.
基金supported by the National Key R&D Program of China(No.2016YFB1200203)the National Natural Science Foundation of China(Nos.41427806 and 61273233)
文摘Deep neural networks have been successfully applied to numerous machine learning tasks because of their impressive feature abstraction capabilities.However,conventional deep networks assume that the training and test data are sampled from the same distribution,and this assumption is often violated in real-world scenarios.To address the domain shift or data bias problems,we introduce layer-wise domain correction(LDC),a new unsupervised domain adaptation algorithm which adapts an existing deep network through additive correction layers spaced throughout the network.Through the additive layers,the representations of source and target domains can be perfectly aligned.The corrections that are trained via maximum mean discrepancy,adapt to the target domain while increasing the representational capacity of the network.LDC requires no target labels,achieves state-of-the-art performance across several adaptation benchmarks,and requires significantly less training time than existing adaptation methods.
