Land use change in rural China since the 1980s, induced by institution reforms, urbanization, industrialization and population increase, has received more attention. However, case studies on how institution reforms af...Land use change in rural China since the 1980s, induced by institution reforms, urbanization, industrialization and population increase, has received more attention. However, case studies on how institution reforms affect farmers＇ livelihood strategies and drive land use change are scarce. By means of cropland plots investigations and interviews with farmers, this study examines livelihood strategy change and land use change in Danzam Village of Jinchuan County in the upper Dadu River watershed, eastern Tibetan Plateau, China. The results show that, during the collective system period, as surplus labor forces could not be transferred to the secondary and tertiary industries, they had to choose agricultural involution as their livelihood strategy, then the farmers had to produce more grains by land reclamation, increasing multiple cropping index, improving input of labor, fertilizer, pesticide and adopting advanced agricultural techniques. During the household responsibility system period, as labors being transferred to the secondary and tertiary industries, farmers chose livelihood diversification strategy. Therefore, labor input to grain planting was greatly reduced, which drove the transformation of grain to horticulture, vegetable or wasteland and decrease of multiple cropping index. This study provides a new insight into understanding linkages among institution reforms, livelihood strategy of smallholders and land use change in rural China.展开更多
Deep neural networks have been successfully applied to numerous machine learning tasks because of their impressive feature abstraction capabilities.However,conventional deep networks assume that the training and test ...Deep neural networks have been successfully applied to numerous machine learning tasks because of their impressive feature abstraction capabilities.However,conventional deep networks assume that the training and test data are sampled from the same distribution,and this assumption is often violated in real-world scenarios.To address the domain shift or data bias problems,we introduce layer-wise domain correction(LDC),a new unsupervised domain adaptation algorithm which adapts an existing deep network through additive correction layers spaced throughout the network.Through the additive layers,the representations of source and target domains can be perfectly aligned.The corrections that are trained via maximum mean discrepancy,adapt to the target domain while increasing the representational capacity of the network.LDC requires no target labels,achieves state-of-the-art performance across several adaptation benchmarks,and requires significantly less training time than existing adaptation methods.展开更多
基金Under the auspices of National Natural Science Foundation of China (No. 40601006, 40471009)National Basic Rsearch Program of China (No. 2005CB422006)
文摘Land use change in rural China since the 1980s, induced by institution reforms, urbanization, industrialization and population increase, has received more attention. However, case studies on how institution reforms affect farmers＇ livelihood strategies and drive land use change are scarce. By means of cropland plots investigations and interviews with farmers, this study examines livelihood strategy change and land use change in Danzam Village of Jinchuan County in the upper Dadu River watershed, eastern Tibetan Plateau, China. The results show that, during the collective system period, as surplus labor forces could not be transferred to the secondary and tertiary industries, they had to choose agricultural involution as their livelihood strategy, then the farmers had to produce more grains by land reclamation, increasing multiple cropping index, improving input of labor, fertilizer, pesticide and adopting advanced agricultural techniques. During the household responsibility system period, as labors being transferred to the secondary and tertiary industries, farmers chose livelihood diversification strategy. Therefore, labor input to grain planting was greatly reduced, which drove the transformation of grain to horticulture, vegetable or wasteland and decrease of multiple cropping index. This study provides a new insight into understanding linkages among institution reforms, livelihood strategy of smallholders and land use change in rural China.
基金supported by the National Key R&D Program of China(No.2016YFB1200203)the National Natural Science Foundation of China(Nos.41427806 and 61273233)
文摘Deep neural networks have been successfully applied to numerous machine learning tasks because of their impressive feature abstraction capabilities.However,conventional deep networks assume that the training and test data are sampled from the same distribution,and this assumption is often violated in real-world scenarios.To address the domain shift or data bias problems,we introduce layer-wise domain correction(LDC),a new unsupervised domain adaptation algorithm which adapts an existing deep network through additive correction layers spaced throughout the network.Through the additive layers,the representations of source and target domains can be perfectly aligned.The corrections that are trained via maximum mean discrepancy,adapt to the target domain while increasing the representational capacity of the network.LDC requires no target labels,achieves state-of-the-art performance across several adaptation benchmarks,and requires significantly less training time than existing adaptation methods.
