ID 原文 译文
39526 边缘计算资源的地理分散性、异构性以及对性能、能耗、费用、稳定性等的需求，增加了优化调度的复杂性。通过介绍边缘计算和物联网、云计算协同的系统模型，给出优化的指标、调度模型及其求解算法，包括精确算法、启发式方法及智能优化方法等， It becomes more difficult due to the geographic separated and heterogeneous features of edge computing resource as well as the requirements of performance, energy consumption, cost and stability and introduces the system models of edge computing, IoT and cloud computing, presents the optimization metrics, scheduling models and solution optimization, including exact algorithms, heuristic methods and intelligent optimization algorithms.
39527 归纳典型应用案例，指出有待进一步研究的内容和方向，有助于促进边缘计算的发展。 In addition, typical application cases, and points out the further research contents and directions are provided to promote the development of edge computing.
39528 鉴于羊群行为的复杂性和重要性，基于消费者行为的最新研究成果，以多智能体建模的研究方法仿真并对比分析不同类型复杂网络对羊群效应现象的影响。 Herd behavior is valuable but complicated. The impact of different type complex network on herd effect is analyzed by multi-agent modeling and simulation. In the model, herd behavior of decision-maker is modeled based on the advanced evidences from empirical study.
39529 模型中，决策者的羊群行为采用最新的实证研究成果。通过仿真，不仅验证了数学建模分析能够得到的结论，还因过程仿真而能够发现羊群效应现象的时间特性。 Through simulation, not only the conclusions drawn from traditional mathematical modeling are verified, but also the whole evolving process of herd effect is inspected and the evolving speed related features can be used to evaluate performances of herd effect.
39530 通过对常见的小世界网络和无标度网络环境下的羊群行为仿真对比，发现网络结构是决定羊群效应和厚尾效应显著性的关键因素之一。 More important, simulation results show the network structure of different type complex networks is one of the key factors for herd effect and heavy-tail effect.
39531 例如，“棋盘式”网格网络和E-R随机网络环境下的羊群效应更加明显，小世界网络环境下的羊群效应相比之下并不明显，而无标度网络环境会出现厚尾现象等。 For instance, homogenous networks, i.e. regular grid network and E-R random network present a significant herd effect, small world network shows less herd effect, while heavy-tail effect can be found in scale free network environment.
39532 另外，仿真结果表明复杂的消费者行为与复杂的网络结构叠加后，其对羊群效应现象的影响反而并不显著。 When combining with heterogeneous random network, the influence of non-linear herd behavior is not obvious for herd effect.
39533 卷积神经网络(Convolutional Neural Network，CNN)在合成孔径雷达(Synthetic Aperture Radar，SAR)图像目标识别领域得到广泛应用。 Convolutional neural networks have been widely used in the field of synthetic aperture radar image target recognition.
39534 在Le Net-5神经网络模型的基础上，提出了跨卷积网络特征融合的SAR图像识别方法。 Based on the LeNet-5 neural network model, a SAR image target recognition method are initialized across convolution network feature fusion is proposed.
39535 利用MNIST手写数据对LeNet-5网络参数进行初始化，提取SAR图像的深层特征和浅层特征，对浅层特征进行主成分分析以得到关键类别信息， The LeNet-5 network parameters on the basis of MNIST handwritten data. The deep and shallow features of the SAR image are extracted, and the principal component analysis on the shallow features is performed to obtain key category information.