提出了联合的半参数概率图模型用于学习非正态分布异构数据的条件相关性。进一步针对光滑变化的异构数据,给出了联合的动态半参数概率图模型。将基于非参排序的相关矩阵估计方法与融合图套索方法相结合,提出了半参数融合图套索方法估计...提出了联合的半参数概率图模型用于学习非正态分布异构数据的条件相关性。进一步针对光滑变化的异构数据,给出了联合的动态半参数概率图模型。将基于非参排序的相关矩阵估计方法与融合图套索方法相结合,提出了半参数融合图套索方法估计上述两类联合图模型。特别针对动态半参数图模型,提出了一种新的核光滑Kendall’s tau相关矩阵。由于放宽了正态分布假设,使得该模型比当前联合的高斯图模型更灵活。由于采用了基于非参排序的相关矩阵估计方法,使得该模型更加鲁棒。利用有效的交替方向乘子法(alternating direction method of multipliers,ADMM)对这些模型进行求解。最后,在一些人工数据与真实数据上,如脑影像数据及股票交易数据,验证了该模型的有效性。展开更多
This paper describes a novel method of online composite shape recognition interms of the relevance feedback technology to capture a user's intentions incrementally, and adynamic user modeling method to adapt to va...This paper describes a novel method of online composite shape recognition interms of the relevance feedback technology to capture a user's intentions incrementally, and adynamic user modeling method to adapt to various users' styles. First, the relevance feedback isadapted to refine the recognition results and reduce the ambiguity incrementally based on theestablishment of a feature-based vector model of a user's sketches. Secondly, a dynamic usermodeling is introduced to model the user's sketching habits based on recording and analyzinghistorical information incrementally. A model-based matching strategy is also employed in the methodto recognize sketches dynamically. Experiments prove that the proposed method is both effective andefficient.展开更多
We consider an epidemical model within soclally interacting mobile individuals to study the behaviors of steady states of epidemic propagation in 2D networks. Using mean-field approximation and large scale simulations...We consider an epidemical model within soclally interacting mobile individuals to study the behaviors of steady states of epidemic propagation in 2D networks. Using mean-field approximation and large scale simulations, we recover the usual epidemic behavior with critical thresholds δc and pc below which infectious disease dies out. For the population density δ far above δc it is found that there is linear relationship between contact rate λ and the population density δ in the main. At the same time, the result obtained from mean-field approximation is compared with our numerical result, and it is found that these two results are similar by and large but not completely the same.展开更多
This work presents a formalized description of information and communicative interac- tions of individuals on the basis of the communicative field （CF） method. It also contains explication of the possibility to crea...This work presents a formalized description of information and communicative interac- tions of individuals on the basis of the communicative field （CF） method. It also contains explication of the possibility to create model of information and communication inter- actions, which is able to illustrate both interactions between two and more individuals. Methods and approaches which are suggested in this paper can correctly simulate the processes of distortion and generation of information images （IIs） with information and communication social interaction. Expansion and addition of IIs theory in terms of the transmission of information between individuals allows us to speak about the space of IIs. This space helps to explain a number of characteristic phenomena of human thinking.展开更多
文摘提出了联合的半参数概率图模型用于学习非正态分布异构数据的条件相关性。进一步针对光滑变化的异构数据,给出了联合的动态半参数概率图模型。将基于非参排序的相关矩阵估计方法与融合图套索方法相结合,提出了半参数融合图套索方法估计上述两类联合图模型。特别针对动态半参数图模型,提出了一种新的核光滑Kendall’s tau相关矩阵。由于放宽了正态分布假设,使得该模型比当前联合的高斯图模型更灵活。由于采用了基于非参排序的相关矩阵估计方法,使得该模型更加鲁棒。利用有效的交替方向乘子法(alternating direction method of multipliers,ADMM)对这些模型进行求解。最后,在一些人工数据与真实数据上,如脑影像数据及股票交易数据,验证了该模型的有效性。
文摘This paper describes a novel method of online composite shape recognition interms of the relevance feedback technology to capture a user's intentions incrementally, and adynamic user modeling method to adapt to various users' styles. First, the relevance feedback isadapted to refine the recognition results and reduce the ambiguity incrementally based on theestablishment of a feature-based vector model of a user's sketches. Secondly, a dynamic usermodeling is introduced to model the user's sketching habits based on recording and analyzinghistorical information incrementally. A model-based matching strategy is also employed in the methodto recognize sketches dynamically. Experiments prove that the proposed method is both effective andefficient.
基金The project supported by National Natural Science Foundation of China under Grant No. 50272022 and the Sunshine Foundation of Wuhan City under Grant No. 20045006071-40
文摘We consider an epidemical model within soclally interacting mobile individuals to study the behaviors of steady states of epidemic propagation in 2D networks. Using mean-field approximation and large scale simulations, we recover the usual epidemic behavior with critical thresholds δc and pc below which infectious disease dies out. For the population density δ far above δc it is found that there is linear relationship between contact rate λ and the population density δ in the main. At the same time, the result obtained from mean-field approximation is compared with our numerical result, and it is found that these two results are similar by and large but not completely the same.
基金Research was partially supported by a Grant from the Russian Science Foundation （Project No. 15-18-00047）. This research was partially supported by Grants from the Board President of the Russian Federation （Project MK-7165.2015.6）.
文摘This work presents a formalized description of information and communicative interac- tions of individuals on the basis of the communicative field （CF） method. It also contains explication of the possibility to create model of information and communication inter- actions, which is able to illustrate both interactions between two and more individuals. Methods and approaches which are suggested in this paper can correctly simulate the processes of distortion and generation of information images （IIs） with information and communication social interaction. Expansion and addition of IIs theory in terms of the transmission of information between individuals allows us to speak about the space of IIs. This space helps to explain a number of characteristic phenomena of human thinking.
