Abstract In this study, aiming at the multi-attribute decision-making problem with incomplete and uncertain attribute weight information and attribute value of interval numbers, a grey target decision-making model of interval numbers based on positive and negative clouts is proposed. Firstly, in this model, the linear transformation operator of interval number is used to normalize the original decision information, and the positive and negative clouts of interval number are designed. Secondly, after the space projection distance between each scheme and the positive and negative clouts is considered comprehensively, the off-target distance is taken as the basis of vector analysis in space to obtain a new comprehensive off-target distance. The existing interval number grey target decision-making model ignores the important influence of interval distribution and the correlation between the attributes in scheme evaluation, and there are some fuzzy errors when setting the weight of attributes. In order to solve the above problems, this paper combined with the uncertainty analysis of the attribute weights, a goal programming is constructed for the objective function based on the comprehensive off-target distance minimization to solve the attribute weight vector, and finally determine the order of the scheme. Finally, the feasibility and effectiveness of the proposed grey target decision model are verified by an example of venture capital projects. Compared with traditional models, the improved model fully considers the characteristics of interval data and the correlation between the attributes.

To support large-scale intelligent applications, distributed machine learning based on JointCloud is an intuitive solution scheme. However, the distributed machine learning is difficult to train due to that the corresponding optimization solver algorithms converge slowly, which highly demand on computing and memory resources. To overcome the challenges, we propose a computing framework for L-BFGS optimization algorithm based on variance reduction method, which can utilize a fixed big learning rate to linearly accelerate the convergence speed. To validate our claims, we have conducted several experiments on multiple classical datasets. Experimental results show that the computing framework accelerate the training process of solver and obtain accurate results for machine learning algorithms.

Consistency and responsiveness are two important factors in providing the sense of reality in Distributed Virtual Environment (DVE). However, it is not easy to optimize both aspects because of the trade-off between these two factors. As a result, most existing consistency maintenance methods ignored the responsiveness requirements, or just assumed a simple responsiveness requirement model which cannot meet the real need of DVE systems. In this paper, we first present a new responsiveness requirement model. The model can describe requirement satisfaction situation of each node. Base on this model, we propose a responsiveness requirement based consistency method. The method can adjust the utilization of time resource according to the requirements of different nodes and improve the overall responsiveness performance by at least 20%. Therefore, it provides a good support to increase the applicability of DVE systems.