系统科学与复杂网络研究

从模型的角度出发，分析系统科学所讨论的几类模型，总结研究这些模型的主要概念和主要方法；依据当前复杂网络研究的实际，提出系统科学和复杂网络结合研究的思路、具体方法、应注意的问题。     

Establishing formal state space models via quantization forquantum control systems

1.INTRODUCTION Quantumtheoryisoneofthemostimportantachieve mentsinthe20thcentury,anditdisclosesthestruc tures,characteristicsandmotionlawsofmatterinthe microcosmicfieldandleadsusfromthemacroscopical worldintomicrosystems.Aseriesofnewphenomena whicharedifferentfromclassicalsystemshavebeen discovered,suchasquantumentanglement,quantum coherenceanduncertainty.Meanwhile,quantum theoryandquantummethodsarealsowidelyusedin differentfieldssuchaschemicalreaction,geneticen gineering,atomicphysicsandi…     

迅速发展的复杂网络研究与面临的挑战

近年来复杂网络的小世界效应、无标度特性和超家族的发现,大大地激起了国内外研究复杂网络的热潮.本文简介了若干研究进展,讨论了当前整个复杂网络研究面临的挑战性问题.     

卷积神经网络声学模型的结构优化和加速计算

将卷积神经网络(convolutional neural networks,CNN)声学模型应用于中文大词表连续电话语音识别任务中,分析了卷积层数、滤波器参数等变量对CNN模型性能的影响,最终在中文电话语音识别测试中,CNN模型相比传统的全连接神经网络模型取得了识别字错误率1.2％的下降.由于卷积结构的复杂性,常规的神经网络加速方法如定点量化和SSE指令加速等方法对卷积运算的加速效率较低.针对这种情况,对卷积结构进行了优化,提出了2种卷积矢量化方法:权值矩阵矢量化和输入矩阵矢量化对卷积运算进行改善.结果表明,输入矩阵矢量化方法的加速效率更高,结合激活函数后移的策略,使得卷积运算速度提升了8.9倍.     

基于折扣高斯粒子滤波器的桥梁可靠性动态预测

桥梁健康监测(BHM)系统在长期运营中积累了大量信息,如何利用这些信息动态预测结构可靠性已成为BHM领域的关键科学问题之一.为合理预测桥梁的动态可靠性,应用BHM系统日常监测的极值应力数据,建立带有最优折扣因子的动态线性模型,结合高斯粒子滤波器给出折扣高斯粒子滤波器预测算法,分别对日常监测极值应力的一步向前预测分布参数和状态变量后验分布参数进行修正预测,并基于此,采用一次二阶矩(FOSM)方法预测桥梁的动态可靠性,结合桥梁实测数据对所提方法进行了验证分析,为桥梁预防性养护维修决策提供理论基础.     

中国洛阳牡丹文化节发展现状与对策研究——以第31届中国洛阳牡丹文化节为例

中国洛阳牡丹文化节的前身是始于1983年的洛阳牡丹花会,迄今已经连续举办了32届,尤其自2011年升格为国家级节会以来更是取得重大突破,其对地区经济社会发展的促进作用也日益凸显。然而,升级后中国洛阳牡丹文化节仍存在若干问题,突出表现在宣传工作、文化活动、节会瓶颈、旅游消费等方面。未来完善和提升牡丹文化节的对策主要有:加强创新、完善节会宣传模式;广泛参与、彰显民众主体地位;凝聚合力、突破节会发展瓶颈;积极引导、挖掘公众消费潜力等。     

Forecasting VaR models under Different Volatility Processes and Distributions of Return Innovations

This paper provides clear‐cut evidence that the out‐of‐sample VaR (value‐at‐risk) forecasting performance of alternative parametric volatility models, like EGARCH (exponential general autoregressive conditional heteroskedasticity) or GARCH, and Markov regime‐switching models, can be considerably improved if they are combined with skewed distributions of asset return innovations. The performance of these models is found to be similar to that of the EVT (extreme value theory) approach. The performance of the latter approach can also be improved if asset return innovations are assumed to be skewed distributed. The performance of the Markov regime‐switching model is considerably improved if this model allows for EGARCH effects, for all different volatility regimes considered. Copyright © 2014 John Wiley & Sons, Ltd.     

Forward Rates,Monetary Policy and the Economic Cycle

The short end of the yield curve incorporates essential information to forecast central banks' decisions, but in a biased manner. This article proposes a new method to forecast the Fed and the European Central Bank's decision rate by correcting the swap rates for their cyclical economic premium, using an affine term structure model. The corrected yields offer a higher out‐of‐sample forecasting power than the yields themselves. They also deliver forecasts that are either comparable or better than those obtained with a factor‐augmented vector autoregressive model, underlining the fact that yields are likely to contain at least as much information regarding monetary policy as a dataset composed of economic data series. Copyright © 2015 John Wiley & Sons, Ltd.     

Long Memory of Financial Time Series and Hidden Markov Models with Time‐Varying Parameters

Hidden Markov models are often used to model daily returns and to infer the hidden state of financial markets. Previous studies have found that the estimated models change over time, but the implications of the time‐varying behavior have not been thoroughly examined. This paper presents an adaptive estimation approach that allows for the parameters of the estimated models to be time varying. It is shown that a two‐state Gaussian hidden Markov model with time‐varying parameters is able to reproduce the long memory of squared daily returns that was previously believed to be the most difficult fact to reproduce with a hidden Markov model. Capturing the time‐varying behavior of the parameters also leads to improved one‐step density forecasts. Finally, it is shown that the forecasting performance of the estimated models can be further improved using local smoothing to forecast the parameter variations. Copyright © 2016 John Wiley & Sons, Ltd.     

Making the abstract concrete: The role of norms and values in experimental modeling

Experimental modeling is the construction of theoretical models hand in hand with experimental activity. As explained in Section 1, experimental modeling starts with claims about phenomena that use abstract concepts, concepts whose conditions of realization are not yet specified; and it ends with a concrete model of the phenomenon, a model that can be tested against data. This paper argues that this process from abstract concepts to concrete models involves judgments of relevance, which are irreducibly normative. In Section 2, we show, on the basis of several case studies, how these judgments contribute to the determination of the conditions of realization of the abstract concepts and, at the same time, of the quantities that characterize the phenomenon under study. Then, in Section 3, we compare this view on modeling with other approaches that also have acknowledged the role of relevance judgments in science. To conclude, in Section 4, we discuss the possibility of a plurality of relevance judgments and introduce a distinction between locally and generally relevant factors.