基于卷积神经网络的列车检测算法研究

列车检测作为列车自动驾驶的核心技术,可以有效地降低列车追尾等事故造成的人身危险和财产损失。为实现精准的列车检测,选用改进的卷积神经网络(PVANET)对输入图像进行特征提取,在此基础上,采用候选区域网络,从生成的特征图里滑动搜索,判断出图像中可能为列车的区域位置,并进一步采用快速区域卷积神经网络对每个候选区域进行分类,计算出其所属类别的置信度,同时精确定位列车。经验证,该方法适应范围广、鲁棒性高,可以有效地检测不同环境光强及不同朝向的列车,保障列车安全,为列车自动驾驶及辅助驾驶提供安全保障。     

A hybrid machine learning model for forecasting a billing period’s peak electric load days

Many models have been studied for forecasting the peak electric load, but studies focusing on forecasting peak electric load days for a billing period are scarce. This focus is highly relevant to consumers, as their electricity costs are determined based not only on total consumption, but also on the peak load required during a period. Forecasting these peak days accurately allows demand response actions to be planned and executed efficiently in order to mitigate these peaks and their associated costs. We propose a hybrid model based on ARIMA, logistic regression and artificial neural networks models. This hybrid model evaluates the individual results of these statistical and machine learning models in order to forecast whether a given day will be a peak load day for the billing period. The proposed model predicted 70% (40/57) of actual peak load days accurately and revealed potential savings of approximately USD $80,000 for an American university during a one-year testing period.     

效率视角下"一带一路"沿线国家创新影响力研究

从效率差异视角构建国家创新影响力评价方法，揭示“一带一路”沿线国家创新要素流动倾向性以促进交流合作。运用DEA模型测算各国创新效率，根据DEA原理识别国家间的认可关系，构建创新网络并用节点特征向量中心度表示各国创新影响力，分析创新影响力类型、创新网络特征及创新投入冗余率与产出不足率。结果发现，各国创新效率存在显著差距;各国分别处于创新网络中心、核心和边缘位置，国家创新影响力依次递减;按照创新要素相对强度，沿线国家可分为6种创新类型;最后，提出沿线国家间开展创新交流与合作的必要性和途径。     

Predicting bank insolvencies using machine learning techniques

Proactively monitoring and assessing the economic health of financial institutions has always been the cornerstone of supervisory authorities. In this work, we employ a series of modeling techniques to predict bank insolvencies on a sample of US-based financial institutions. Our empirical results indicate that the method of Random Forests (RF) has a superior out-of-sample and out-of-time predictive performance, with Neural Networks also performing almost equally well as RF in out-of-time samples. These conclusions are drawn not only by comparison with broadly used bank failure models, such as Logistic, but also by comparison with other advanced machine learning techniques. Furthermore, our results illustrate that in the CAMELS evaluation framework, metrics related to earnings and capital constitute the factors with higher marginal contribution to the prediction of bank failures. Finally, we assess the generalization of our model by providing a case study to a sample of major European banks.     

Interbank tiering and money center banks

This paper provides evidence that interbank markets are tiered rather than flat, in the sense that most banks do not lend to each other directly but through money center banks acting as intermediaries. We capture the concept of tiering by developing a core-periphery model, and devise a procedure for fitting the model to real-world networks. Using Bundesbank data on bilateral interbank exposures among 2000 banks from 1999 to 2012, we find strong evidence of tiering in the German banking system. This extent of tiering is unlikely to arise in standard random networks. Indeed, we show that bank specialization and balance sheet variables predict how banks position themselves in the interbank market. This link provides a promising avenue for understanding the formation of financial networks.     

演化视角下全产业链创新网络结构及核心企业行为对创新绩效的影响——以苹果公司网络为例

从理论上分析了全产业链创新网络结构特点以及核心企业行为模式与创新绩效的关系,提出相关假设;从开放性的2个维度,即开放广度和开放深度量化核心企业开放式创新的行为模式。依据偏最小二乘法,将结构和行为模式的表征变量作为自变量,创新绩效的4个表征变量作为因变量,进行交叉检验;最后选取1个主成分进行回归分析,检验理论假设。     

The emergence of multi-sided platform MNEs: Internalization theory and networks

The rise of the digital economy provides firms across the globe with unique business opportunities. Companies such as Facebook, Alibaba, and Uber are competing in a new multi-sided platform world; the primary focus of these firms, from their inception, is to provide digital infrastructure, information and technology—intangible assets that enable direct interaction or value creation across platforms by linking different user group and complementors, often at the international level. Building on data drawn from multinational multisided Platform corporations (MMPCs) operating in China, we combine insights from internalization theory and network effects in understanding the value creation of such firms. We explore the boundaries of these new “breed” of MNEs in exploiting firm-specific advantages (FSAs) and in creating new knowledge between headquarters and subsidiaries. The findings suggest that internalization theory needs to shift its focus from the ‘boundaries of the firm’ to the ‘boundaries of the local network’. By integrating their internal and external networks of knowledge in adapting their business models in host markets, this new breed of MNEs is more likely than the traditional one to gain a sustainable competitive advantage in the new information age.     

多层级视角下京津冀科技创新政策府际关系研究

府际关系协调有利于实现创新资源、要素的有效汇聚与整合，进而提升政策有效性。以京津冀为例，构建中央-地方多层级府际关系模型，分析中央政策主体合作网络特征与政策扩散特征、地方政策主体合作网络特征与政策执行特征的相关性。研究发现，中央政府部门间形成了以科技部为核心的核心—边缘网络结构特征；北京市政府部门间呈现出以中关村科技管理委员会为核心的核心-边缘网络结构特征；天津市政府部门间呈现出以天津市科学技术委员会为核心的星型网络结构特征；河北省则没有形成稳定的网络关系。中央政策主体合作网络中的度中心性与政策扩散广度具有相关性；地方政策主体合作网络中介中心性、度中心性与政策执行强度具有相关性。     

Co‐movements in commodity prices: a note based on network analysis

This article analyses co‐movements in a wide group of commodity prices during the time period 1992–2010. Our methodological approach is based on the correlation matrix and the networks inside. Through this approach we are able to summarize global interaction and interdependence, capturing the existing heterogeneity in the degrees of synchronization between commodity prices. Our results produce two main findings: (a) we do not observe a persistent increase in the degree of co‐movement of the commodity prices in our time sample, however from mid‐2008 to the end of 2009 co‐movements almost doubled when compared with the average correlation; (b) we observe three groups of commodities which have exhibited similar price dynamics (metals, oil and grains, and oilseeds) and which have increased their degree of co‐movement during the sampled period.     

Alternative farming and collective goals: Towards a powerful relationships for future food policies

Based on the significant amount of literature on the concept of spatial, ecological, and social embeddedness in the Alternative Food Networks (AFNs) research, this study describes the main dynamics related to the involvement of non-conventional farmers in their collective ideas and actions.With a comparison between two European regions: Sardinia and the Community of Madrid, the paper identifies different styles of behaviour among organic producers, whether or not supervised by a control body, and highlights the differences and similarities regarding their ideas about how alternative agriculture has an impact on the environment as well as society. Eco-economies and ecopreneurship are also considered. The aim is to see if, and to what extent, these ideas have a collective character, going beyond the limits of the single farm to reach a regional scale and, thus, if there is a basis for the development of future food-related planning policies. Through the information provided by semi-structured interviews, farmers have been divided into categories according to their degree of embeddedness, described through the main themes that emerged during the interviews. Every farmer has been included in only one category according to his or her main preference, which does not imply the ab-sence of a positive attitude towards other categories.The study aims to contribute to the understanding of how AFN values and methods can boost the improvement of biodiversity and landscape conservation through collective actions which have the power to boost and develop rural initiative at a regional scale.