Neural networks for contract bridge bidding

The objective of this study is to explore the possibility of capturing the reasoning process used in bidding a hand in a bridge game by an artificial neural network. We show that a multilayer feedforward neural network can be trained to learn to make an opening bid with a new hand. The game of bridge, like many other games used in artificial intelligence, can easily be represented in a machine. But, unlike most games used in artificial intelligence, bridge uses subtle reasoning over and above the agreed conventional system, to make a bid from the pattern of a given hand. Although it is difficult for a player to spell out the precise reasoning process he uses, we find that a neural network can indeed capture it. We demonstrate the results for the case of one-level opening bids, and discuss the need for a hierarchical architecture to deal with bids at all levels.     

夏季面漆起泡的预防

面漆漆膜在夏季潮湿天气下成膜后3—5天，漆膜表面易形成米粒状的漆泡，针对漆泡的现象，寻找其形成的机理，根据起泡的机理采取预防面漆起泡的措施。     

米塔尔钢铁集团的全球并购发展战略

概述了米塔尔的并购经过和并购特点,着重探讨了米塔尔的管理模式、人员情况、钢铁厂布局、生产经营、主要战略伙伴、研发机构及其生产、采购与销售策略。系统地分析了米塔尔在中国的发展,指出米塔尔在全球钢铁业整合中处于领先地位,米塔尔来中国投资看中的是中国的原料和市场,且运作步伐明显加快,会对中国钢铁企业构成很大的威胁。     

Lake-Level Change and Water Balance Analysis at Lake Qinghai,West China during Recent Decades

Lake Qinghai, the largest saline lake with an area of 4,260 km² (2000) and average depth of 21 m (1985) in West China, has experienced severe decline in water level in recent decades. This study aimed to investigate water balance of the lake and identify the causes for the decline in lake level. There was a 3.35-m decline in water level with an average decreasing rate of 8.0 cm year⁻¹ between 1959 and 2000. The lake water balance showed that mean annual precipitation between 1959 and 2000 over the lake was 357 ± 10 mm, evaporation was 924 ± 10 mm, surface runoff water inflow was 348 ± 21 mm, groundwater inflow was 138 mm ± 9 and the change in lake level was −80 ± 31 mm. The variation of lake level was highly positively correlated to surface runoff and precipitation and negatively to evaporation, the correlation coefficients were 0.89, 0.81 and −0.66, respectively. Water consumption by human activities accounts for 1% of the evaporation loss of the lake, implying that water consumption by human activities has little effect on lake level decline. Most dramatic decline in lake level occurred in the warm and dry years, and moderate decline in the cold and dry years, and relatively slight decline in the warm and wet years, therefore, the trend of cold/warm and dry climate in recent decades may be the main reasons for the decline in lake level.     

人工神经网络技术在板料激光弯曲中的应用

基于人工神经网络基本理论,建立在板料激光弯曲中预测材料表面最高温度、弯曲角度的BP网络模型。借助于MATLAB仿真软件中的神经网络工具箱作为开发平台,将试验样本数据和经过试验验证的数值计算结果作为补充的样本数据用于BP网络的训练,利用训练好的BP网络对非线性的样本数据规律进行拟合,实现激光加工工艺参数的优化,为实际生产和加工提供有效的依据。     

RSFM网络及其在信号滤波中的应用

介绍了一种适用于模式识别的新型神经网络模型--局部有监督特征映射（Reglonal Supervised Feature Mapping,RSFM）网络,描述了该网络的拓扑结构和学习算法,研究了网络的基本性能,最后将其应用到了信号滤波中.理论研究和仿真实验表明,该网络结构简单、算法简洁,收敛速度快、识别精度高,适用于需要大样本训练、随机干扰严重的复杂模式的分类与识别.     

提高油田集输系统可靠性的研究

应用现代可靠性理论,从设计、工艺、技术等方面对油田集输系统进行了可靠性探讨,提出了增强系统可靠性的措施。     

山区建筑的结构设计注意点及探讨——永安一中教学楼设计心得

本文通过工程实例分析了山区建筑的结构设计特点,并提出了结构设计中应注意的问题。     

降雨空间分布模式识别

以广东省内186个站点的年降雨量分布为基础，把模式识别的聚类分析理论方法引入降雨空间特征分类研究中，进行降雨特征空间分布模式识别，以预测无雨量站点的降雨模式，从而分析水资源的空间分布模式，以便制订有效的水资源利用方案。结果表明，广东省年降雨量在空间上的分布可划分为6类模式；同时基于BP人工神经网络算法，建立了以高程、气温、蒸发量为主要参数的可判别和预测无雨量测站区域的降雨空间分布模式所属类别的模型，经过模型的学习和验证，效果较好。     

美国西雅图先锋广场历史保护区的形成

考察美国西雅图先锋广场历史保护区形成的过程,可以看出城市历史文化遗存的保护是经济与历史文化合力作用的结果,公众参与在城市历史文化保护工作中发挥着重要的作用。我国城市经济高速发展,保护与发展的矛盾冲突还没得到很好解决,先锋广场的例子对我们具有一定的参考价值。