In this paper, we investigate Galton-Watson branching processes in random environments. In the case where the environmental process is a Markov chain which is positive recurrent or has a transition matrix Q (θ,α) su...In this paper, we investigate Galton-Watson branching processes in random environments. In the case where the environmental process is a Markov chain which is positive recurrent or has a transition matrix Q (θ,α) such that sup_θ Q (θ,α)> 0 for some α, we prove that the model has the asymptotic behavior being similar to that of Galton-Watson branching processes. In other case where the environments are non-stationary independent, the sufficient conditions are obtained for certain extinction and uncertain extinction for the model.展开更多
A novel deep neural network compression model for airport object detection has been presented.This novel model aims at disadvantages of deep neural network,i.e.the complexity of the model and the great cost of calcula...A novel deep neural network compression model for airport object detection has been presented.This novel model aims at disadvantages of deep neural network,i.e.the complexity of the model and the great cost of calculation.According to the requirement of airport object detection,the model obtains temporal and spatial semantic rules from the uncompressed model.These spatial semantic rules are added to the model after parameter compression to assist the detection.The rules can improve the accuracy of the detection model in order to make up for the loss caused by parameter compression.The experiments show that the effect of the novel compression detection model is no worse than that of the uncompressed original model.Even some of the original model false detection can be eliminated through the prior knowledge.展开更多
We consider the small value probability of supercritical continuous state branching processes with immigration. From Pinsky （1972） it is known that under regularity condition on the branching mechanism and immigrati...We consider the small value probability of supercritical continuous state branching processes with immigration. From Pinsky （1972） it is known that under regularity condition on the branching mechanism and immigration mechanism, the normalized population size converges to a non-degenerate finite and positive limit PV as t tends to infinity. We provide sharp estimate on asymptotic behavior of P（W≤ε〈） as ε→ 0＋ by studying the Laplace transform of W. Without immigration, we also give a simpler proof for the small value probability in the non-subordinator case via the prolific backbone decomposition.展开更多
文摘In this paper, we investigate Galton-Watson branching processes in random environments. In the case where the environmental process is a Markov chain which is positive recurrent or has a transition matrix Q (θ,α) such that sup_θ Q (θ,α)> 0 for some α, we prove that the model has the asymptotic behavior being similar to that of Galton-Watson branching processes. In other case where the environments are non-stationary independent, the sufficient conditions are obtained for certain extinction and uncertain extinction for the model.
文摘A novel deep neural network compression model for airport object detection has been presented.This novel model aims at disadvantages of deep neural network,i.e.the complexity of the model and the great cost of calculation.According to the requirement of airport object detection,the model obtains temporal and spatial semantic rules from the uncompressed model.These spatial semantic rules are added to the model after parameter compression to assist the detection.The rules can improve the accuracy of the detection model in order to make up for the loss caused by parameter compression.The experiments show that the effect of the novel compression detection model is no worse than that of the uncompressed original model.Even some of the original model false detection can be eliminated through the prior knowledge.
基金supported by National Science Foundation of US (Grant Nos. DMS-0805929 and DMS-1106938)National Natural Science Foundation of China (Grant Nos. 10928103,10971003 and 11128101)+1 种基金Specialized Research Fund for the Doctoral Program of Higher Education of Chinathe Fundamental Research Funds for the Central Universities
文摘We consider the small value probability of supercritical continuous state branching processes with immigration. From Pinsky （1972） it is known that under regularity condition on the branching mechanism and immigration mechanism, the normalized population size converges to a non-degenerate finite and positive limit PV as t tends to infinity. We provide sharp estimate on asymptotic behavior of P（W≤ε〈） as ε→ 0＋ by studying the Laplace transform of W. Without immigration, we also give a simpler proof for the small value probability in the non-subordinator case via the prolific backbone decomposition.
