一种基于PVM的并行BP神经网络

通过分析单计算机的BP神经网络学习和执行算法,提出了使用PVM构造多计算机的并行神经网络。该并行神经网络的实现灵活应用于高可靠性和大规模数据的分析和处理中。同时,该BP神经网络的并行设计与实现,可广泛应用于其他神经网络模型的并行计算机实现。

The influence and chronological uncertainties of the 8.2 ka cooling event on continental climate records in China

IEECAS SKLLQG

Elemental carbon record of paleofire history on the Chinese Loess Plateau during the last 420 ka and its response to environmental and climate changes

IEECAS SKLLQG

考古生物指标记录的中国西北地区5000a BP水稻遗存

IEECAS SKLLQG

巴丹吉林沙漠查格勒布剖面微量元素反映的150ka BP以来的气候变化

IEECAS SKLLQG

Summer monsoon intensity controls C-4/C-3 plant abundance during the last 35 ka in the Chinese Loess Plateau: Carbon isotope evidence from bulk organic matter and individual leaf waxes

IEECAS SKLLQG

Did an extensive forest ever develop on the Chinese Loess Plateau during the past 130 ka?: a test using soil carbon isotopic signatures

IEECAS SKLLQG

巴丹吉林查格勒布剖面记录的150ka BP以来的常量化学元素波动

IEECAS SKLLQG

2．5ka BP以来额济纳绿洲沙丘的粒度特征及其反映的沉积过程

IEECAS SKLLQG