神经网络应用

神经网络的应用是神经网络研究的三大主要领域之一。神经网络的应用与其所具备的能力是紧密相关的。本文中,我们从介绍神经网络的基本能力开始,概括地介绍了当前的神经网络主要应用领域。并从神经网络的能力出发,讨论了神经网络的潜在的应用领域。     

Ada网络文件传输系统的实现

本文介绍了一个在NETBIOS支持下用Ada的任务机制实现的网络上任意两节点间的文件全双工传输系统。传输时,文件以数据报的分组形式发送,由Ada 的任务特性来管理收、发进程,实现后台全双工操作。本文还简要地介绍了实现中的一些技巧。     

神经计算机容错能力分析

容错能力是计算机(传统计算机或神经计算机)能力的一个极重要的方面。本文从研究传统计算机容错技术和人脑容错特性出发,着重研究了神经计算机的容错能力,并得出了对神经计算机设计极有价值的结论。     

具有分流型抑制机制的视网膜细胞神经网络模型及其应用

视觉细胞之间存在的侧抑制机制能够对接收到的视觉信息进行选择性的提取。运用数学方法模拟视觉细胞之间存在的这种生理机制,提出了一种具有分流型抑制机制的视网膜细胞神经网络模型(SIRCNN),分析了该模型的稳定性,给出了等价电路图,并且基于此模型提出了一种边缘检测算法。将该算法应用于图像边缘检测中,试验结果表明,这种以生物视觉感知机理为基础的神经计算方法可有效地增强图像反差,检测图像边缘。由于细胞神经网络具有高速并行运算、便于硬件实现等特点,因此这种方法在图像实时处理中也具有很大的潜力和应用前景。     

无线自组织网络跨层攻击检测博弈模型

无线自组织网络中的跨层攻击具有比单层攻击更强的隐蔽性、更好的攻击效果或更低的攻击成本.为了检测无线自组织网络中的跨层攻击,提出了一种基于博弈论的攻击检测模型.由于攻击不可避免地会对各协议层的参数造成影响,因此模型从协议层攻防博弈的角度,建立起相应的策略矩阵和支付矩阵,并通过均衡分析得到该模型的混合策略纳什均衡解.仿真结...     

Disruption management in production planning

We study the problem of recovering a production plan after a disruption, where the disruption may be caused by incidents such as power failure, market change, machine breakdown, supply shortage, worker no‐show, and others. The new recovery plan we seek after has to not only suit the changed environment brought about by the disruption, but also be close to the initial plan so as not to cause too much customer unsatisfaction or inconvenience for current‐stage and downstream operations. For the general‐cost case, we propose a dynamic programming method for the problem. For the convex‐cost case, a general problem which involves both cost and demand disruptions can be solved by considering the cost disruption first and then the demand disruption. We find that a pure demand disruption is easy to handle; and for a pure cost disruption, we propose a greedy method which is provably efficient. Our computational studies also reveal insights that will be helpful to managing disruptions in production planning. © 2005 Wiley Periodicals, Inc. Naval Research Logistics, 2005.     

An approximate planning model for distributed computing networks

We develop an approximate planning model for a distributed computing network in which a control system oversees the assignment of information flows and tasks to a pool of shared computers, and describe several optimization applications using the model. We assume that the computers are multithreaded, and have differing architectures leading to varying and inconsistent processing rates. The model is based on a discrete‐time, continuous flow model developed by Graves [Oper Res 34 (1986), 522–533] which provides the steady‐state moments of production and work‐in‐queue quantities. We make several extensions to Graves' model to represent distributed computing networks. First, we approximately model control rules that are nonlinear functions of the work‐in‐queue at multiple stations through a linearization approach. Second, we introduce an additional noise term on production and show its use in modeling the discretization of jobs. Third, we model groups of heterogeneous computers as aggregate, “virtual computing cells” that process multiple tasks simultaneously, using a judiciously selected control rule. © 2005 Wiley Periodicals, Inc. Naval Research Logistics, 2005.     

基于FNNs的闭式循环柴油机配氧前馈控制策略研究

配氧控制是闭式循环柴油机(CCD)系统的关键技术之一,为改善其动态特性,在 PID反馈控制的基础上,设计了基于模糊神经网络(FNNs)模型的前馈控制器,并采用 PID反馈控制输出信号作为网络训练的误差信号,使模糊神经网络逐步具有前馈补偿能力,从而能够有效对负荷扰动进行及时补偿.仿真结果表明,采用FNNs前馈控制器后,可以有效改善氧气控制的动态特性,并且具有快速的学习速度和很强的适应能力.     

长向量处理器高效RNN推理方法

模型深度的不断增加和处理序列长度的不一致对循环神经网络在不同处理器上的性能优化提出巨大挑战。针对自主研制的长向量处理器FT-M7032,实现了一个高效的循环神经网络加速引擎。该引擎采用行优先矩阵向量乘算法和数据感知的多核并行方式,提高矩阵向量乘的计算效率;采用两级内核融合优化方法降低临时数据传输的开销;采用手写汇编优化多种算子,进一步挖掘长向量处理器的性能潜力。实验表明,长向量处理器循环神经网络推理引擎可获得较高性能,相较于多核ARM CPU以及Intel Golden CPU,类循环神经网络模型长短记忆网络可获得最高62.68倍和3.12倍的性能加速。     

Small world,big guns: globalization,interstate security networks and conventional weapons imports

With the burgeoning influence of emerging markets in Asia, a tectonic shift is taking place in the global security landscape. Asian states are concomitantly arming as their economic clout grows. In light of these developments, security analysts would benefit from a formal means of placing these arms acquisitions in a structural context. Are arms acquisitions on par with the expectations of Asian states, given their structural dispositions, or are recent acquisitions beyond anticipated levels? By using a dynamic panel regression of 187 states from 1950 to 2011, this research predicts arms import volume using the degree of interstate arms linkages, the size of a state's military, and its level of economic development. The technique offers analysts a formal means of distinguishing orthodox behavior in importing conventional weapons from extraneous security motivations. The article concludes by generating near-term forecasts of Asian arms imports and discussing the implications of the technique.