Through analysis, it is believed that major reasons for default risks in operation of small amount credit include low management level and vacancy of normative system, vacancy of risk sharing mechanism, rating distort...Through analysis, it is believed that major reasons for default risks in operation of small amount credit include low management level and vacancy of normative system, vacancy of risk sharing mechanism, rating distortion due to imperfect credit investigation system, and uncertainty of borrower's credit. On the basis of these, static and dynamic models are established to analyze the prevention mechanism for default risk in small amount credit. It is concluded that we must establish a restriction mechanism during operation of small amount credit as long as three values increase, namely, N (potential loss of bad credit record due to farmers' default), Q (probability of successful recovery by small amount credit institution), and S (cost of small amount credit institution punishing farmers after successful recovery). Finally, following countermeasures and suggestions are put forward: perfect laws and regulations and credit reward and punishment mechanism for risk management of small amount credit; bring into play proper function of loan officer in small amount credit practice; widely promote rural "Group Credit Union" system.展开更多
针对生成对抗网络(GAN)需要大量训练数据及点云数据稀缺且获取难度大的问题,提出一种基于少量数据集的三维点云生成模型。该模型首先通过重采样和水平旋转的方法实现数据增强,使第一级网络能够生成具有多样性的低分辨率点云;其次通过确...针对生成对抗网络(GAN)需要大量训练数据及点云数据稀缺且获取难度大的问题,提出一种基于少量数据集的三维点云生成模型。该模型首先通过重采样和水平旋转的方法实现数据增强,使第一级网络能够生成具有多样性的低分辨率点云;其次通过确保低分辨率点云与高分辨率点云之间的对应关系,实现点云的超分辨率生成;最后实现生成具有多样性的高分辨率点云。实验结果表明,在ShapeNet Part(ShapeNet Part Segmentation Dataset)数据集上,该模型的JS散度相较于Tree-GAN的JS散度下降了0.416,证明其性能优于Tree-GAN。展开更多
基金Supported by the Project for Decision-making Bidding of Henan Provincial Government (2011B546)
文摘Through analysis, it is believed that major reasons for default risks in operation of small amount credit include low management level and vacancy of normative system, vacancy of risk sharing mechanism, rating distortion due to imperfect credit investigation system, and uncertainty of borrower's credit. On the basis of these, static and dynamic models are established to analyze the prevention mechanism for default risk in small amount credit. It is concluded that we must establish a restriction mechanism during operation of small amount credit as long as three values increase, namely, N (potential loss of bad credit record due to farmers' default), Q (probability of successful recovery by small amount credit institution), and S (cost of small amount credit institution punishing farmers after successful recovery). Finally, following countermeasures and suggestions are put forward: perfect laws and regulations and credit reward and punishment mechanism for risk management of small amount credit; bring into play proper function of loan officer in small amount credit practice; widely promote rural "Group Credit Union" system.
文摘针对生成对抗网络(GAN)需要大量训练数据及点云数据稀缺且获取难度大的问题,提出一种基于少量数据集的三维点云生成模型。该模型首先通过重采样和水平旋转的方法实现数据增强,使第一级网络能够生成具有多样性的低分辨率点云;其次通过确保低分辨率点云与高分辨率点云之间的对应关系,实现点云的超分辨率生成;最后实现生成具有多样性的高分辨率点云。实验结果表明,在ShapeNet Part(ShapeNet Part Segmentation Dataset)数据集上,该模型的JS散度相较于Tree-GAN的JS散度下降了0.416,证明其性能优于Tree-GAN。
