Subexponentially Growing Hilbert Space and Nonconcentrating Distributions in a Constrained Spin Model

Motivated by recent experiments with two-component Bose–Einstein condensates, we study fully-connected spin models subject to an additional constraint. The constraint is responsible for the Hilbert space dimension to scale only linearly with the system size. We discuss the unconventional statistical physical and thermodynamic properties of such a system, in particular the absence of concentration of the underlying probability distributions. As a consequence, expectation values are less suitable to characterize such systems, and full distribution functions are required instead. Sharp signatures of phase transitions do not occur in such a setting, but transitions from singly peaked to doubly peaked distribution functions of an “order parameter” may be present.     

PAC-Bayes Bounds on Variational Tempered Posteriors for Markov Models

Datasets displaying temporal dependencies abound in science and engineering applications, with Markov models representing a simplified and popular view of the temporal dependence structure. In this paper, we consider Bayesian settings that place prior distributions over the parameters of the transition kernel of a Markov model, and seek to characterize the resulting, typically intractable, posterior distributions. We present a Probably Approximately Correct (PAC)-Bayesian analysis of variational Bayes (VB) approximations to tempered Bayesian posterior distributions, bounding the model risk of the VB approximations. Tempered posteriors are known to be robust to model misspecification, and their variational approximations do not suffer the usual problems of over confident approximations. Our results tie the risk bounds to the mixing and ergodic properties of the Markov data generating model. We illustrate the PAC-Bayes bounds through a number of example Markov models, and also consider the situation where the Markov model is misspecified.     

A Bayesian Approach to the Estimation of Parameters and Their Interdependencies in Environmental Modeling

We present a case study for Bayesian analysis and proper representation of distributions and dependence among parameters when calibrating process-oriented environmental models. A simple water quality model for the Elbe River (Germany) is referred to as an example, but the approach is applicable to a wide range of environmental models with time-series output. Model parameters are estimated by Bayesian inference via Markov Chain Monte Carlo (MCMC) sampling. While the best-fit solution matches usual least-squares model calibration (with a penalty term for excessive parameter values), the Bayesian approach has the advantage of yielding a joint probability distribution for parameters. This posterior distribution encompasses all possible parameter combinations that produce a simulation output that fits observed data within measurement and modeling uncertainty. Bayesian inference further permits the introduction of prior knowledge, e.g., positivity of certain parameters. The estimated distribution shows to which extent model parameters are controlled by observations through the process of inference, highlighting issues that cannot be settled unless more information becomes available. An interactive interface enables tracking for how ranges of parameter values that are consistent with observations change during the process of a step-by-step assignment of fixed parameter values. Based on an initial analysis of the posterior via an undirected Gaussian graphical model, a directed Bayesian network (BN) is constructed. The BN transparently conveys information on the interdependence of parameters after calibration. Finally, a strategy to reduce the number of expensive model runs in MCMC sampling for the presented purpose is introduced based on a newly developed variant of delayed acceptance sampling with a Gaussian process surrogate and linear dimensionality reduction to support function-valued outputs.     

基于雷达海杂波的区域性非均匀蒸发波导反演方法

由于不同海域上空气象条件的不同, 海上蒸发波导在大尺度海面上空发生时通常是区域性非均匀的, 这一特性使得该环境中的电波传播特性相对于水平均匀的蒸发波导环境情况而明显不同, 因此, 进行区域性非均匀的蒸发波导探测反演对正确预测电波传播特性及提高雷达系统的工作性能具有重要的意义. 考虑到实际应用中蒸发波导信息获取手段的多样性, 将中尺度数值气象模式MM5预报的区域性蒸发波导修正折射率剖面作为先验信息, 提出了一种含该先验信息的区域性非均匀蒸发波导的雷达海杂波后验概率估计模型. 该模型使用主分量分析法对蒸发波导的水平非均匀性进行参数化建模, 然后通过贝叶斯理论将修正折射率剖面参数的先验概率分布、后验概率分布和似然函数联系起来, 利用雷达海杂波实现蒸发波导剖面参数的最大后验概率估计反演. 通过我国东海海域的实际区域性非均匀蒸发波导反演测试, 表明该模型能够以更高的精度实现区域性非均匀蒸发波导的反演.     

Maximum entropy approach to statistical inference for an ocean acoustic waveguide

A conditional probability distribution suitable for estimating the statistical properties of ocean seabed parameter values inferred from acoustic measurements is derived from a maximum entropy principle. The specification of the expectation value for an error function constrains the maximization of an entropy functional. This constraint determines the sensitivity factor (β) to the error function of the resulting probability distribution, which is a canonical form that provides a conservative estimate of the uncertainty of the parameter values. From the conditional distribution, marginal distributions for individual parameters can be determined from integration over the other parameters. The approach is an alternative to obtaining the posterior probability distribution without an intermediary determination of the likelihood function followed by an application of Bayes' rule. In this paper the expectation value that specifies the constraint is determined from the values of the error function for the model solutions obtained from a sparse number of data samples. The method is applied to ocean acoustic measurements taken on the New Jersey continental shelf. The marginal probability distribution for the values of the sound speed ratio at the surface of the seabed and the source levels of a towed source are examined for different geoacoustic model representations.     

利用Bayesian-MCMC方法从雷达回波反演海洋波导

应用贝叶斯-蒙特卡罗（Bayesian-MCMC）方法将海洋波导参数的先验信息描述为先验概率密度,结合雷达回波资料（电磁波传播损耗）,得到待反演海洋波导参数的后验概率密度,用马尔可夫链蒙特卡罗（MCMC）-Gibbs采样器采样后验概率密度分布,并用样本最大似然估计值作为对海洋波导参数分布的估计.数值实验结果表明,该方法对先验信息进行了有效利用,反演精度高于遗传算法的反演精度.该方法较为充分利用先验信息,得到解的概率分布,即解的不确定性分析,这在实际应用中有一定的参考价值. 关键词： 波导 电磁波传播损耗 贝叶斯-蒙特卡罗 概率分布     

Brownian excursions on combs

In this paper we use techniques from Ito excursion theory to analyze Brownian motion on generalized combs. Ito excursion theory is a little-known area of probability theory and we therefore present a brief introduction for the uninitiated. A general method for analyzing transport along the backbone of the comb is demonstrated and the specific case of a comb whose teeth are scaling branching trees is examined. We then present a recursive method for evaluating the distribution of the first passage times on hierarchical combs.     

A recursive method for calculating the total number of spanning trees and its applications in self-similar small-world scale-free network models

The problem of determining and calculating the number of spanning trees of any finite graph (model) is a great challenge, and has been studied in various fields, such as discrete applied mathematics, theoretical computer science, physics, chemistry and the like. In this paper, firstly, thank to lots of real-life systems and artificial networks built by all kinds of functions and combinations among some simpler and smaller elements (components), we discuss some helpful network-operation, including link-operation and merge-operation, to design more realistic and complicated network models. Secondly, we present a method for computing the total number of spanning trees. As an accessible example, we apply this method to space of trees and cycles respectively, and our results suggest that it is indeed a better one for such models. In order to reflect more widely practical applications and potentially theoretical significance, we study the enumerating method in some existing scale-free network models. On the other hand, we set up a class of new models displaying scale-free feature, that is to say, following P(k) ~ k^?γ, where γ is the degree exponent. Based on detailed calculation, the degree exponent γ of our deterministic scale-free models satisfies γ > 3. In the rest of our discussions, we not only calculate analytically the solutions of average path length, which indicates our models have small-world property being prevailing in amounts of complex systems, but also derive the number of spanning trees by means of the recursive method described in this paper, which clarifies our method is convenient to research these models.     

Lattice models of advection-diffusion

We present a synthesis of theoretical results concerning the probability distribution of the concentration of a passive tracer subject to both diffusion and to advection by a spatially smooth time-dependent flow. The freely decaying case is contrasted with the equilibrium case. A computationally efficient model of advection-diffusion on a lattice is introduced, and used to test and probe the limits of the theoretical ideas. It is shown that the probability distribution for the freely decaying case has fat tails, which have slower than exponential decay. The additively forced case has a Gaussian core and exponential tails, in full conformance with prior theoretical expectations. An analysis of the magnitude and implications of temporal fluctuations of the conditional diffusion and dissipation is presented, showing the importance of these fluctuations in governing the shape of the tails. Some results concerning the probability distribution of dissipation, and concerning the spatial scaling properties of concentration fluctuation, are also presented. Though the lattice model is applied only to smooth flow in the present work, it is readily applicable to problems involving rough flow, and to chemically reacting tracers. (c) 2000 American Institute of Physics.     

Continuous probability distributions from finite data

Recent approaches to the problem of inferring a continuous probability distribution from a finite set of data have used a scalar field theory for the form of the prior probability distribution. This paper presents a more general form for the prior distribution that has a geometrical interpretation and can improve the specificity of likely solutions. It is also demonstrated that a numerical sampling of the posterior probability distribution can be used as an alternative to a histogram for visualization and to make probabilistic inferences from the data.     

Bayesian image processing in magnetic resonance imaging.

In the past several years, image processing techniques based on Bayesian models have received considerable attention. In our earlier work, we developed a novel Bayesian approach which was primarily aimed at the processing and reconstruction of images in positron emission tomography. In this paper, we describe how the technique has been adopted to process magnetic resonance images in order to reduce noise and artifacts, thereby improving image quality. In this framework, the image is assumed to be a statistical variable whose posterior probability density conditional on the observed image is modeled by the product of the likelihood function of the observed data with a prior density based our prior knowledge. A Gibbs random field incorporating local continuity information and with edge-detection capability is used as the prior model. Based on the formalism of the posterior density, we can compute an estimate of the image using an iterative technique. We have implemented this technique and applied it to phantom and clinical images. Our results indicate that the approach works reasonably well for reducing noise, enhancing edges, and removing ringing artifact.     

Subordinated α-stable Ornstein–Uhlenbeck process as a tool for financial data description

The classical financial models are based on the standard Brownian diffusion-type processes. However, in the exhibition of some real market data (like interest or exchange rates) we observe characteristic periods of constant values. Moreover, in the case of financial data, the assumption of normality is often unsatisfied. In such cases the popular Vasi?ek model, that is a mathematical system describing the evolution of interest rates based on the Ornstein–Uhlenbeck process, seems not to be applicable. Therefore, we propose an alternative approach based on a combination of the popular Ornstein–Uhlenbeck process with a stable distribution and subdiffusion systems that demonstrate such characteristic behavior. The probability density function of the proposed process can be described by a Fokker–Planck type equation and therefore it can be examined as an extension of the basic Ornstein–Uhlenbeck model. In this paper, we propose the parameters’ estimation method and calibrate the subordinated Vasi?ek model to the interest rate data.     

Analyzing open-source software systems as complex networks

Software systems represent one of the most complex man-made artifacts. Understanding the structure of software systems can provide useful insights into software engineering efforts and can potentially help the development of complex system models applicable to other domains. In this paper, we analyze one of the most popular open-source Linux meta packages/distributions called the Gentoo Linux. In our analysis, we model software packages as nodes and dependencies among them as edges. Our empirical results show that the resulting Gentoo network cannot be easily explained by existing complex network models. This in turn motivates our research in developing two new network growth models in which a new node is connected to an old node with the probability that depends not only on the degree but also on the “age” of the old node. Through computational and empirical studies, we demonstrate that our models have better explanatory power than the existing ones. In an effort to further explore the properties of these new models, we also present some related analytical results.     

Some exact results on the ultrametric overlap distribution in mean field spin glass models (I)

The mean field spin glass model is analyzed by a combination of exact methods and a powerful Ansatz. The method exploited is general, and can be applied to others disordered mean field models such as, e.g., neural networks.It is well known that the probability measure of overlaps among replicas carries the whole physical content of these models. A functional order parameter of Parisi type is introduced by rigorous methods, according to previous works by F. Guerra. By the Ansatz that the functional order parameter is the correct order parameter of the model, we explicitly find the full overlap distribution. The physical interpretation of the functional order parameter is obtained, and ultrametricity of overlaps is derived as a natural consequence of a branching diffusion process.It is shown by explicit construction that ultrametricity of the 3-replicas overlap distribution together with the Ghirlanda-Guerra relations determines the distribution of overlaps among s replicas, for any s, in terms of the one-overlap distribution.     

Convergence of posterior probabilities in the Bayesian inference strategy

The formalism of operational statistics, a generalized approach to probability and statistics, provides a setting within which inference strategies can be studied with great clarity. This paper is concerned with the asymptotic behavior of the Bayesian inference strategy in this setting. We consider a sequence of posterior distributions, obtained from a prior as a result of successive conditionings by the events of an admissible sequence. We identify certain statistical hypotheses whose limiting posterior probabilities converge to one. We describe these hypotheses, and show that when the prior is vague, they contain those probability models which represent the long-run relative frequencies of occurrence for the events in the sequence.     

Heterogenous mean-field analysis of a generalized voter-like model on networks

We propose a generalized framework for the study of voter models in complex networks at the heterogeneous mean-field (HMF) level that (i) yields a unified picture for existing copy/invasion processes and (ii) allows for the introduction of further heterogeneity through degree-selectivity rules. In the context of the HMF approximation, our model is capable of providing straightforward estimates for central quantities such as the exit probability and the consensus/fixation time, based on the statistical properties of the complex network alone. The HMF approach has the advantage of being readily applicable also in those cases in which exact solutions are difficult to work out. Finally, the unified formalism allows one to understand previously proposed voter-like processes as simple limits of the generalized model.     

Exact results for the Barabási model of human dynamics

Human activity patterns display a bursty dynamics with interevent times following a heavy tailed distribution. This behavior has been recently shown to be rooted in the fact that humans assign their active tasks different priorities, a process that can be modeled as a priority queueing system [A.-L. Barabási, Nature (London) 435, 207 (2005)]. In this Letter we obtain exact results for the Barabási model with two tasks, calculating the priority and waiting time distribution of active tasks. We demonstrate that the model has a singular behavior in the extremal dynamics limit, when the highest priority task is selected first. We find that independently of the selection protocol, the average waiting time is smaller or equal to the number of active tasks, and discuss the asymptotic behavior of the waiting time distribution. These results have important implications for understanding complex systems with extremal dynamics.     

Some exact results on the ultrametric overlap distribution in mean field spin glass models (I)

The mean field spin glass model is analyzed by a combination of exact methods and a simple Ansatz. The method exploited is general, and can be applied to others disordered mean field models such as, e.g., neural networks. It is well known that the probability measure of overlaps among replicas carries the whole physical content of these models. A functional order parameter of Parisi type is introduced by rigorous methods, according to previous works by F. Guerra. By the Ansatz that the functional order parameter is the correct order parameter of the model, we explicitly find the full overlap distribution. The physical interpretation of the functional order parameter is obtained, and ultrametricity of overlaps is derived as a natural consequence of a branching diffusion process. It is shown by explicit construction that ultrametricity of the 3-replicas overlap distribution together with the Ghirlanda-Guerra relations determines the distribution of overlaps among s replicas, for any s, in terms of the one-overlap distribution. Received 14 February 2000     

基于Contourlet变换的遥感图像去噪新算法

提出了一个新的有效的基于Contourlet变换的遥感图像去噪方法。对有噪图像进行Contourlet分解;对Contourlet变换系数引入一个几何先验模型,结合噪声和有用信号的条件分布进行贝叶斯估计,得到每一系数作为有用信号的后验概率,以之作为修正因子修正小波萎缩因子;对重构图像进行递归循环运算处理。仿真实验结果表明,去噪后图像去除了常见的伪吉布斯现象,峰值信噪比提高了1~2 dB。     

Topological properties of phylogenetic trees in evolutionary models

The extent to which evolutionary processes affect the shape of phylogenetic trees is an important open question. Analyses of small trees seem to detect non-trivial asymmetries which are usually ascribed to the presence of correlations in speciation rates. Many models used to construct phylogenetic trees have an algorithmic nature and are rarely biologically grounded. In this article, we analyze the topological properties of phylogenetic trees generated by different evolutionary models (populations of RNA sequences and a simple model with inheritance and mutation) and compare them with the trees produced by known uncorrelated models as the backward coalescent, paying special attention to large trees. Our results demonstrate that evolutionary parameters as mutation rate or selection pressure have a weak influence on the scaling behavior of the trees, while the size of phylogenies strongly affects measured scaling exponents. Within statistical errors, the topological properties of phylogenies generated by evolutionary models are compatible with those measured in balanced, uncorrelated trees.