Observer Reliability and Human Inference

This paper presents a formal analysis of the problem of determining the inferential impact of the information in a composite report from a collection of unreliable observers or sensors. Each sensor reports one of a finite number of possible states of a data system linked probabilistically with an ``objective system' whose condition is to be inferred from the data state. The principal assumptions are that the sensors do not ``collaborate' in making their reports and that their reports are conditioned only by the existing data state and not by the actual, unobservable state of the objective system. Use of the notion of conditional independence to express these assumptions gives the analytic expressions a tractable form which sheds light on various inference issues. The paper also briefly discusses current empirical research on the question of how well people actually adjust the impact of inferential evidence to correspond to the unreliability of the sources of information.     

用贝叶斯统计推断进行系统可靠性分析的方法

贝叶斯分析是综合了人为主观判断和试验数据的一种推断方法,根据获得的累计失效时间数据,采用该方法对串联系统可靠性特征量进行了分析。     

Bayesian Inference for Linear Dynamic Models With Dirichlet Process Mixtures

Using Kalman techniques, it is possible to perform optimal estimation in linear Gaussian state-space models. Here, we address the case where the noise probability density functions are of unknown functional form. A flexible Bayesian nonparametric noise model based on Dirichlet process mixtures is introduced. Efficient Markov chain Monte Carlo and sequential Monte Carlo methods are then developed to perform optimal batch and sequential estimation in such contexts. The algorithms are applied to blind deconvolution and change point detection. Experimental results on synthetic and real data demonstrate the efficiency of this approach in various contexts.     

Bayesian Inference on Multiscale Models for Poisson Intensity Estimation: Applications to Photon-Limited Image Denoising

We present an improved statistical model for analyzing Poisson processes, with applications to photon-limited imaging. We build on previous work, adopting a multiscale representation of the Poisson process in which the ratios of the underlying Poisson intensities (rates) in adjacent scales are modeled as mixtures of conjugate parametric distributions. Our main contributions include: 1) a rigorous and robust regularized expectation-maximization (EM) algorithm for maximum-likelihood estimation of the rate-ratio density parameters directly from the noisy observed Poisson data (counts); 2) extension of the method to work under a multiscale hidden Markov tree model (HMT) which couples the mixture label assignments in consecutive scales, thus modeling interscale coefficient dependencies in the vicinity of image edges; 3) exploration of a 2-D recursive quad-tree image representation, involving Dirichlet-mixture rate-ratio densities, instead of the conventional separable binary-tree image representation involving beta-mixture rate-ratio densities; and 4) a novel multiscale image representation, which we term Poisson-Haar decomposition, that better models the image edge structure, thus yielding improved performance. Experimental results on standard images with artificially simulated Poisson noise and on real photon-limited images demonstrate the effectiveness of the proposed techniques.      

A Critical Review of Human Performance Reliability Predictive Methods

Similarities and differences among 22 methods of quantitatively predicting operator and technician performance are described. Emphasis has been given to eight methods most fully developed and most likely to be used by system engineers. Two general techniques are employed: analysis of historical data and computer-simulation of behavioral processes. No general purpose methodology is available; each method deals with some types of tasks and systems more efficiently than others. In general, simulation-based methods are more powerful than nonsimulation methods. Most methods output probability estimates of successful task/system performance and completion time, but are relatively insensitive to equipment design parameters, manpower selection and training needs. With only one exception no operability method utilizes a formal data base as input and in most cases the parameters these input data describe are not specifically indicated. For most methods validation and/or system application data are either lacking or incomplete.     

定时截尾可靠性鉴定试验中参数MTBF的统计推断

用统计学中样本空间排序法对定时截尾可靠性鉴定试验方案中平均故障间隔时间（MT—BF）的统计推断方法进行了理论推导,并给出了置信下限和置信上限系数的表达式,对GJB899—1990中的表达式提出了修订建议。     

Bayes网络与关系模型的集成:概率关系模型

对象关系模型和Bayes网络分别是关系理论和概率理论两个不同领域中最重要的模型，它们首次集成于本文引入的概率关系模型中，作为新型的概率模型，概率关系模型不仅继承了Bayes网络的大部分优点，而且关系特征和对象的概念使它能有效地克服Bayes网络在许多方面的不足，而成为对复杂系统模建的理想工具，是对Bayes网络的重要创新；作为新型的关系模型，概率关系模型也是对传统关系模型的重要创新，具备概率特征的对象关系模型有了处理不确定性问题的能力。概率关系模型的创建对复杂智能信息系统开发研究有有着特别重要的意义，本文首先评述Bayes网络和对象关系模型，然后在此基础上引入概率关系模型。     

MOS器件开启电压值的贝叶斯统计推断

在实践中准确地测定和表征出MOS器件样管的开启电压,对于大规模集成电路的设计以及将MOS器件作为分立器件的在电路应用,均是至关重要的。文章采用贝叶斯统计推断工具用于器件输入曲线的处理,从中提取出与器件开启有关的更为精准和深入的信息。建立合适的分层模型,应用基于马尔科夫链蒙特卡罗(MCMC)算法的最大后验估计(MAP),求取目标量的后验分布。这类算法为目前概率与统计领域的最高级算法。将该先进算法引入到IC领域来分析处理所获取的大数据是后摩尔时代的一个发展方向。     

最大期望模拟退火的贝叶斯变分推理算法

针对贝叶斯变分推理收敛精度低和搜索过程中易陷入局部最优的问题,该文基于模拟退火理论(SA)和最大期望理论(EM),考虑变分推理过程中初始先验对最终结果的影响和变分自由能的优化效率问题,构建了双重EM模型学习变分参数的初始先验,以降低初始先验的敏感性,同时构建逆温度参数改进变分自由能函数,使变分自由能在优化过程得到有效控制,并提出一种基于最大期望模拟退火的贝叶斯变分推理算法.该文使用收敛性准则理论分析算法的收敛性,利用所提算法对一个混合高斯分布实例进行实验仿真,实验结果表明该算法具有较优的收敛结果.     

Comparison of Reliability and Testing Models

We were curious about how well various reliability and testing models would compare with respect to prediction accuracy and testing effectiveness. Therefore, we conducted several experiments to evaluate these properties for the following models: fault tree analysis, geometric and binomial statistical models; and reliability growth models: Yamada S Shape Model, Schneidewind Single Parameter Model, and Schneidewind Software Reliability Model. We developed modified versions of the geometric and binomial models that comprise a new contribution to the body of software reliability research. The Yamada model provided the best prediction accuracy for one of the Shuttle's failure data sets. Future research would involve evaluating all models against multiple data sets.      

最大期望模拟退火的贝叶斯变分推理算法

针对贝叶斯变分推理收敛精度低和搜索过程中易陷入局部最优的问题,该文基于模拟退火理论(SA)和最大期望理论(EM),考虑变分推理过程中初始先验对最终结果的影响和变分自由能的优化效率问题,构建了双重EM模型学习变分参数的初始先验,以降低初始先验的敏感性,同时构建逆温度参数改进变分自由能函数,使变分自由能在优化过程得到有效控...     

一种高效的单播网络自适应拓扑推测算法

为提高单播网络拓扑推测的效率和准确性,本文提出一种高效的单播网络自适应拓扑推测算法.该算法利用探测包中的TTL信息作为二分深度优先搜索排序中参考节点的选择依据,有效降低测量过程中所需的背靠背包对数量,提高拓扑推测的效率;通过对现有的深度优先搜索下网络拓扑推测算法原理的分析,给出该算法能够正确推测出整个网络拓扑的充分条件,在此基础上提出一种自适应判定阈值选择方法,有效提高网络链路参数未知情况下的拓扑推测准确性.仿真实验结果表明,本文所提出的算法具备更高的拓扑推测效率和准确性.     

Flowgraph Models in Reliability and Finite Automata

We discuss the interrelationship of two seemingly unrelated subjects: the theory of finite automata, and reliability theory, finite automata, more generally known as generalized transition graphs, are dasiaconvertedpsila to regular expressions by manipulating their pictorial representation, a directed graph, by elimination of its states one-by-one until two states are left, connected by an edge whose label is a regular expression equivalent to the initially given finite automata or generalized transition graph. Flowgraphs are used to represent semi-Markov reliability models. They are directed graphs with edges labeled with expressions of the form pG(s), where p is the probability of transition from node i to node j, say; and G(s) is the transform (Laplace transform, moment generating function, or characteristic function) of the waiting time in i given that the next transition is to j. Usually, transforms of waiting time distributions (e.g. time to first failure) are obtained from these graph representations by applying Mason's Rule (e.g. Huzurbazar, Mason, and Osaki), or, by the Cofactor Rule. In this paper we are concerned with obtaining transforms of waiting times by direct manipulation of the flowgraphs along the lines in finite automata. The goal of the paper is to observe that identical patterns of reasoning are applicable in both fields. This interconnects two apparently unrelated fields of knowledge, an interesting observation for its own sake but also important from a tool & technique point of view.     

An Approach to Reliability in Measurements in India

Importance of the 3-echelon system and its implementation scheme in India based upon one echelon-I Laboratory (Primary Standards), 4 Regional and other local echelon-II laboratories and 18 state and other echelon-III laboratories broadly under the STQC program, Department of Electronics, Government of India, is discussed with emphasis on compatibility and traceability to National/International Standards. Coordination of various activities through Government agency, uniformity of reporting procedures, etc., current programs with different international agencies and steps taken to ensure compatibility in measurements are briefly discussed.     

Reliability Models of NMR Systems

Majority voted redundant systems are widely used. A reliability model is developed and analyzed for N-tuple Modular Redundancy-NMR: (n + 1)-out-of-(2n + 1)-where the units are subject to stuck-at-0, stuck-at-1 or stuck-at-X failures and where failures can occur in a mutually compensatory manner. A reconfiguration of the NMR redundancy, the NMR/Simplex strategy, is proposed and evaluated and its model shown to be included in the general model for the compensated NMR.     

基于可信度的不确定推理辅助定密

定密工作是保密工作的前提和基础,在人工定密的基础上,利用计算机辅助定密人员完成电子文档的定密工作,可以减少人为因素的影响,提高定密的准确度.本文从基于可信度的不确定推理模型入手,提出了一种基于可信度的不确定推理辅助定密方法.该方法利用从保密法规中抽取出的规则构建定密规则库,并通过专家打分的方式对每一条规则设定可信度,计算规则匹配后相关密级的可信度来确定涉密文档所属密级.实验证明,该方法的准确率较高,具有一定的参考价值.