一种基于优化马尔科夫特征的图像篡改盲检测算法

文中主要针对拼接图像篡改检测,提出了一种基于优化马尔科夫特征的盲检测算法.该算法在传统马尔科夫特征的基础上,研究了不同相邻BDCT系数对的关联性对于拼接图像的检测能力,进而设计了一种基于互信息量最大化的加权BDCT系数转移概率特征;同时,通过对所有BDCT系数对进行预分组,降低了算法的计算量以及最终的特征维度.最后,采用支持向量机（SVM）作为分类器,在哥伦比亚大学提供的标准图像拼接库上完成测试,取得了较高的平均检测准确率（91.2％）,优于现有的代表性方法.     

一种基于马尔可夫模型的软件可靠性评估方法

针对目前软件可靠性不易评估的特点,建立一种基于马儿可夫链的软件可靠性评估模型。该模型在软件运行流程图的框架下,利用一定的统计学方法,使用线性代数方法来计算软件运行流程中各个状态的概率,建立软件可靠性评估模型,从而计算软件的可靠性;最后,利用软件设计时的判别准则来判断软件是否符合需求。     

Toward block copolymers from nonliving isospecific single‐site catalytic systems

Ethene/1‐olefin blocky copolymers were obtained through nonliving insertion copolymerizations promoted by an isospecific single site catalyst. Propene or 4‐methyl‐1‐pentene were copolymerized with ethene with metallocenes endowed with different stereospecificity in propene polymerization: (i) aspecific “constrained geometry” half‐sandwich complex, {η¹:η⁵‐([tert‐butyl‐amido)dimethylsilyl](2,3,4,5‐tetramethyl‐1‐cyclopentadienyl)}titanium dichloride [Me₂Si(Me₄Cp)(N‐^tBu)TiCl₂] ( CG ), (ii) moderately isospecific rac‐ethylenebis(indenyl)zirconium dichloride [rac‐(EBI)ZrCl₂] ( EBI ), (iii) slightly more isospecific hydrogenated homologue, rac‐ethylenebis(tetrahydroindenyl)zirconium dichloride [rac‐(EBTHI)ZrCl₂] ( EBTHI ), (iv) highly iso‐specific rac‐[methylenebis(3‐tert‐butyl‐1‐indenyl)]zirconium dichloride [rac‐H₂C‐(3‐^tBuInd)₂ZrCl₂] ( TBI ), (v) most isospecific rac‐[isopropylidene‐bis(3‐tert‐butyl‐cyclopentadienyl)]zirconium dichloride [rac‐Me₂C‐(3‐^tBuCp)₂ZrCl₂] ( TBC ). Copolymerizations were described by a 2^nd order Markovian copolymerization model and data are proposed to correlate the formation of 1‐olefin sequences with catalytic site isospecificity, made by the cooperation of organometallic complex and growing chain. Blocky copolymers were prepared over wide ranges of compositions: with any of the isospecific metallocenes when 4‐methyl‐1‐pentene was the 1‐olefin and only with the highly isospecific ones ( TBI , TBC ) when propene was the comonomer. A penultimate unit effect was observed with TBI as the metallocene, whereas a 1^st order Markov model described the ethene/propene copolymerization from TBC . A moderately isospecific metallocene, such as EBI , is shown to be able to prepare blocky ethene copolymers with 4‐methyl‐1‐pentene. These results pave the way for the synthesis of new ethene based materials. © 2010 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 48: 2063–2075, 2010     

A composite Bayesian hierarchical model of compositional data with zeros

We present an effective approach for modelling compositional data with large concentrations of zeros and several levels of variation, applied to a database of elemental compositions of forensic glass of various use types. The procedure consists of the following: (i) partitioning the data set in subsets characterised by the same pattern of presence/absence of chemical elements and (ii) fitting a Bayesian hierarchical model to the transformed compositions in each data subset. We derive expressions for the posterior predictive probability that newly observed fragments of glass are of a certain use type and for computing the evidential value of glass fragments relating to two competing propositions about their source. The model is assessed using cross‐validation, and it performs well in both the classification and evidence evaluation tasks. Copyright © 2014 John Wiley & Sons, Ltd.     

Strong law of large numbers for countable nonhomogeneous Markov chains

The aim of this paper is to establish a strong law of large numbers for the bivariate functions of countable nonhomogeneous Markov chains under the condition of uniform convergence in the Cesàro sense which differs from my previous results. As corollaries, we generalize one of the Liu and Liu’s results for the univariate functions case and obtain another Shannon–McMillan–Breiman theorem for this Markov chains.     

并行子带HMM最大后验概率自适应非线性类估计算法

目前,自动语音识别(ASR)系统在实验室环境下获得了较高的识别率,但是在实际环境中,由于受到背景噪声和传输信道的影响,系统的识别性能急剧恶化.本文以听觉试验为基础,提出一种新的独立子带并行最大后验概率的非线性类估计算法,用以提高识别系统的鲁棒性.本算法利用多种噪声和识别内容功率谱差异,以及噪声在不同频带上对HMM影响的不同,采用多层感知机(MLP)对噪声环境下最大后验概率进行非线性映射,以减少识别系统由于环境不匹配而导致的识别性能下降.实验表明:该算法性能明显优于最大后验线性回归算法和Sangita提出的子带语音识别算法.     

An envelope theorem and some applications to discounted Markov decision processes

In this paper, an Envelope Theorem (ET) will be established for optimization problems on Euclidean spaces. In general, the Envelope Theorems permit analyzing an optimization problem and giving the solution by means of differentiability techniques. The ET will be presented in two versions. One of them uses concavity assumptions, whereas the other one does not require such kind of assumptions. Thereafter, the ET established will be applied to the Markov Decision Processes (MDPs) on Euclidean spaces, discounted and with infinite horizon. As the first application, several examples (including some economic models) of discounted MDPs for which the et allows to determine the value iteration functions will be presented. This will permit to obtain the corresponding optimal value functions and the optimal policies. As the second application of the ET, it will be proved that under differentiability conditions in the transition law, in the reward function, and the noise of the system, the value function and the optimal policy of the problem are differentiable with respect to the state of the system. Besides, various examples to illustrate these differentiability conditions will be provided. This work was partially supported by Benemérita Universidad Aut ónoma de Puebla (BUAP) under grant VIEP-BUAP 38/EXC/06-G, by Consejo Nacional de Ciencia y Tecnología (CONACYT), and by Evaluation-orientation de la COopération Scientifique (ECOS) under grant CONACyT-ECOS M06-M01.     

Dynamic programming analysis of the TV game “Who wants to be a millionaire?”

This paper uses dynamic programming to investigate when contestants should use lifelines or when they should just stop answering in the TV quiz show ‘Who wants to be a millionaire?’. It obtains the optimal strategies to maximize the expected reward and to maximize the probability of winning a given amount of money.     

Entropy of Hidden Markov Processes via Cycle Expansion

Hidden Markov Processes (HMP) is one of the basic tools of the modern probabilistic modeling. The characterization of their entropy remains however an open problem. Here the entropy of HMP is calculated via the cycle expansion of the zeta-function, a method adopted from the theory of dynamical systems. For a class of HMP this method produces exact results both for the entropy and the moment-generating function. The latter allows to estimate, via the Chernoff bound, the probabilities of large deviations for the HMP. More generally, the method offers a representation of the moment-generating function and of the entropy via convergent series.     

Phase-type distribution of customer relationship with Markovian response and marketing expenditure decision on the customer lifetime value

In recent years, more and more companies have adopted relationship marketing (RM). At the core of RM is the development and maintenance of long-term relationships with valuable customers. In RM, the customer lifetime value (CLV) is the discounted profit streams of a customer across the entire customer life cycle. The CLV plays a key role in customer acquisition and retention decisions. In this paper, we present a general mathematical framework for RM, and introduce a Markov chain model which is appropriate in modeling RM because of its flexibility and probabilistic nature. We also develop the life distribution of the customer relationship, which is a phase-type distribution since it is the distribution of the first arrival stopping state. And we obtain the expectation of the CLV, which is an important statistic for good decision-making. Finally, we illustrate how to find the optimal remarketing policy numerically. The framework developed for RM systems in this paper should be seen as a practical approach to RM where one can directly apply the results of phase-type distribution and expectation of CLV to marketing decisions.