Suchergebnisse

As a human rights researcher with national affections, the author of Human Rights: China's Road tries to understand the underlying logic of human rights situations in China and the progress happening there.The author believes that the idea of human rights protection is unconditionally agreed upon by everyone, but the choice of specific patterns and routes is neither justified nor possible to remain unchanged through all the different concrete scenarios. For the 1.4 billion Chinese people, only they themselves are entitled to determine how they should protect their own human rights, and how to

This Ph. D. work is motivated by the possibility of monitoring the conditions of components of energy systems for their extended and safe use, under proper practice of operation and adequate policies of maintenance. The aim is to develop a Support Vector Regression (SVR)-based framework for predicting time series data under stationary/nonstationary environmental and operational conditions. Single SVR and SVR-based ensemble approaches are developed to tackle the prediction problem based on both small and large datasets. Strategies are proposed for adaptively updating the single SVR and SVR-based ensemble models in the existence of pattern drifts. Comparisons with other online learning approaches for kernel-based modelling are provided with reference to time series data from a critical component in Nuclear Power Plants (NPPs) provided by Electricité de France (EDF). The results show that the proposed approaches achieve comparable prediction results, considering the Mean Squared Error (MSE) and Mean Relative Error (MRE), in much less computation time. Furthermore, by analyzing the geometrical meaning of the Feature Vector Selection (FVS) method proposed in the literature, a novel geometrically interpretable kernel method, named Reduced Rank Kernel Ridge Regression-II (RRKRR-II), is proposed to describe the linear relations between a predicted value and the predicted values of the Feature Vectors (FVs) selected by FVS. Comparisons with several kernel methods on a number of public datasets prove the good prediction accuracy and the easy-of-tuning of the hyperparameters of RRKRR-II. ; Ce travail de thèse est motivée par la possibilité de surveiller l'état des composants de systèmes d'énergie pour leur utilisation prolongée et sécuritaire, conformément à la pratique correcte de fonctionnement et des politiques adéquates de maintenance. La motivation est de développer des méthodes basées sur la régression à vecteurs de support pour la prédiction de données de séries chronologiques dans des conditions environnementales ...

This Ph. D. work is motivated by the possibility of monitoring the conditions of components of energy systems for their extended and safe use, under proper practice of operation and adequate policies of maintenance. The aim is to develop a Support Vector Regression (SVR)-based framework for predicting time series data under stationary/nonstationary environmental and operational conditions. Single SVR and SVR-based ensemble approaches are developed to tackle the prediction problem based on both small and large datasets. Strategies are proposed for adaptively updating the single SVR and SVR-based ensemble models in the existence of pattern drifts. Comparisons with other online learning approaches for kernel-based modelling are provided with reference to time series data from a critical component in Nuclear Power Plants (NPPs) provided by Electricité de France (EDF). The results show that the proposed approaches achieve comparable prediction results, considering the Mean Squared Error (MSE) and Mean Relative Error (MRE), in much less computation time. Furthermore, by analyzing the geometrical meaning of the Feature Vector Selection (FVS) method proposed in the literature, a novel geometrically interpretable kernel method, named Reduced Rank Kernel Ridge Regression-II (RRKRR-II), is proposed to describe the linear relations between a predicted value and the predicted values of the Feature Vectors (FVs) selected by FVS. Comparisons with several kernel methods on a number of public datasets prove the good prediction accuracy and the easy-of-tuning of the hyperparameters of RRKRR-II. ; Ce travail de thèse est motivée par la possibilité de surveiller l'état des composants de systèmes d'énergie pour leur utilisation prolongée et sécuritaire, conformément à la pratique correcte de fonctionnement et des politiques adéquates de maintenance. La motivation est de développer des méthodes basées sur la régression à vecteurs de support pour la prédiction de données de séries chronologiques dans des conditions environnementales et opérationnelles stationnaire/ non-stationnaire. Les simples modèles et les ensembles de modèles à base de SVR sont développés pour attaquer la prédiction basée sur des petits et des grands ensembles de données. Des stratégies sont proposées pour la mise à jour de façon adaptative les simples modèles et les ensembles de modèles à base de SVR au cas du changement de la distribution générant les données. Les comparaisons avec d'autres méthodes d'apprentissage en ligne sont fournies en référence à des données de séries chronologiques d'un composant critique dans les centrales nucléaires fournis par Electricité de France (EDF). Les résultats montrent que les approches proposées permettent d'atteindre des résultats de prédiction comparables compte tenu de l'erreur quadratique moyenne et erreur relative, en beaucoup moins de temps de calcul. Par ailleurs, en analysant le sens géométrique de la méthode de la sélection de vecteurs caractéristiques(FVS) proposé dans la littérature, une nouvelle méthode géométriquement interprétable, nommé Reduced RankKernel Ridge Regression-II (RRKRR-II), est proposée pour décrire les relations linéaires entre un valeur prédite et les valeurs prédites des vecteurs caractéristiques sélectionné par FVS. Les comparaisons avec plusieurs méthodes sur un certain nombre de données publics prouvent la bonne précision de la prédiction et le réglage facile des hyperparamètres de RRKRR-II.

China's Achievements in Poverty Reduction and its World Significance -- People-centered Development Philosophy -- Overall Design of National Poverty Reduction Governance System -- Precise Poverty Reduction Strategy -- Building the Pattern of Poverty Alleviation -- Poverty Alleviation and Poverty Eradication Model -- Poverty Alleviation Should be Combined with Ideological and Intellectual Support -- Poverty Alleviation through Comprehensive Protection -- International Cooperation for Poverty Reduction -- Conception of China's Poverty Reduction Strategy After 2020.

AbstractIn the online retailing, consumers are commonly uncertain about the product's quality and fitness. To resolve these uncertainties, many pure e‐tailers adopt various omnichannel strategies to provide tactile product information for consumers. We build a model to investigate a pure e‐tailer's decision on whether to adopt an omnichannel strategy. Our result indicates that when the cost for each physical store is sufficiently low, the e‐tailer always adopts the omnichannel strategy regardless of the product quality. Moreover, the low‐quality e‐tailer's willingness to adopt the omnichannel strategy is nonmonotonic with the fitness probability when the travel cost factor is high. In contrast, if the cost for each physical store is moderate, the e‐tailer adopts the omnichannel strategy if and only if the product quality is above a threshold. The quality threshold may increase with the fitness probability. Higher fitness probability means a lower return rate and fewer benefits brought by the omnichannel strategy. Thus, the threshold of the quality is increased to guarantee a sufficiently large price increase when choosing the omnichannel strategy. Furthermore, when the cost for each physical store is high, the e‐tailer with a high‐quality product would abandon the omnichannel strategy if the fitness probability is moderate. Finally, we consider the scenarios in which the e‐tailer can endogenously determine the number of physical stores or provide a partial refund policy.