期刊文献+
共找到20篇文章
< 1 >
每页显示 20 50 100
Pricing European Options Based on a Logarithmic Truncated t-Distribution
1
作者 Yingying Cao Xueping Liu +1 位作者 Yiqian Zhao Xuege Han 《Journal of Applied Mathematics and Physics》 2023年第5期1349-1358,共10页
The t-distribution has a “fat tail” feature, which is more suitable than the normal probability density function to describe the distribution characteristics of return on assets. The difficulty of using t-distributi... The t-distribution has a “fat tail” feature, which is more suitable than the normal probability density function to describe the distribution characteristics of return on assets. The difficulty of using t-distribution to price European options is that a fat tail can lead to a deviation in one integral required for option pricing. We use a distribution called logarithmic truncated t-distribution to price European options. A risk neutral valuation method was used to obtain a European option pricing model with logarithmic truncated t-distribution. 展开更多
关键词 Option Pricing Logarithmic Truncated t-distribution Asset Returns Risk-Neutral Valuation Approach
下载PDF
An Extended Bivariate T-Distribution Type Symmetry Model for Square Contingency Tables 被引量:1
2
作者 Kiyotaka Iki Masayuki Okada Sadao Tomizawa 《Open Journal of Statistics》 2018年第2期249-257,共9页
The purpose of this paper is to propose a new model of asymmetry for square contingency tables with ordered categories. The new model may be appropriate for a square contingency table if it is reasonable to assume an ... The purpose of this paper is to propose a new model of asymmetry for square contingency tables with ordered categories. The new model may be appropriate for a square contingency table if it is reasonable to assume an underlying bivariate t-distribution with different marginal variances having any degrees of freedom. As the degrees of freedom becomes larger, the proposed model approaches the extended linear diagonals-parameter symmetry model, which may be appropriate for a square table if it is reasonable to assume an underlying bivariate normal distribution. The simulation study based on bivariate t-distribution is given. An example is given. 展开更多
关键词 BIVARIATE t-distribution SQUARE CONTINGENCY Table SYMMETRY UNDERLYING Distribution
下载PDF
A Multi-Strategy-Improved Northern Goshawk Optimization Algorithm for Global Optimization and Engineering Design
3
作者 Liang Zeng Mai Hu +2 位作者 Chenning Zhang Quan Yuan Shanshan Wang 《Computers, Materials & Continua》 SCIE EI 2024年第7期1677-1709,共33页
Optimization algorithms play a pivotal role in enhancing the performance and efficiency of systems across various scientific and engineering disciplines.To enhance the performance and alleviate the limitations of the ... Optimization algorithms play a pivotal role in enhancing the performance and efficiency of systems across various scientific and engineering disciplines.To enhance the performance and alleviate the limitations of the Northern Goshawk Optimization(NGO)algorithm,particularly its tendency towards premature convergence and entrapment in local optima during function optimization processes,this study introduces an advanced Improved Northern Goshawk Optimization(INGO)algorithm.This algorithm incorporates a multifaceted enhancement strategy to boost operational efficiency.Initially,a tent chaotic map is employed in the initialization phase to generate a diverse initial population,providing high-quality feasible solutions.Subsequently,after the first phase of the NGO’s iterative process,a whale fall strategy is introduced to prevent premature convergence into local optima.This is followed by the integration of T-distributionmutation strategies and the State Transition Algorithm(STA)after the second phase of the NGO,achieving a balanced synergy between the algorithm’s exploitation and exploration.This research evaluates the performance of INGO using 23 benchmark functions alongside the IEEE CEC 2017 benchmark functions,accompanied by a statistical analysis of the results.The experimental outcomes demonstrate INGO’s superior achievements in function optimization tasks.Furthermore,its applicability in solving engineering design problems was verified through simulations on Unmanned Aerial Vehicle(UAV)trajectory planning issues,establishing INGO’s capability in addressing complex optimization challenges. 展开更多
关键词 Northern Goshawk Optimization tent chaotic map t-distribution disturbance state transition algorithm UAV path planning
下载PDF
应用t-SNE算法探讨实验室检查在自身免疫性疾病诊断上临床意义
4
作者 肖瑞平 朱有凯 《中国科技期刊数据库 医药》 2023年第8期59-62,共4页
应用机器学习算法t-SNE(t-Distributed Stochastic Neighbor Embedding)对自身免疫性疾病患者实验室检查数据进行数据分析,探索其中数据结构、数据之间的关系以及在自身免疫性疾病诊断方面的意义。方法 构建以t-SNE为基础的数据分析模型... 应用机器学习算法t-SNE(t-Distributed Stochastic Neighbor Embedding)对自身免疫性疾病患者实验室检查数据进行数据分析,探索其中数据结构、数据之间的关系以及在自身免疫性疾病诊断方面的意义。方法 构建以t-SNE为基础的数据分析模型,以原始实验室检查数据生成的大量高维数据集反复训练模型,确定各种重要参数和实验流程,最终对生成的一系列可视化散点图进行分析,揭示其中包含的信息和知识。结果 本研究建立了可靠性与实用性较强的数据分析模型以及具有临床实践意义的数据分析流程。通过对880例常见自身免疫性疾病病种的数据分析,发现超敏C反应蛋白将所有病例显著地分为两大类;同病种的病例具有明显聚集的数据簇结构,不同病例的数据点有重叠现象;通过比较不同的数据集分析结果,进一步简化了检查项目组合。结论 采用本研究建立的数据分析模型,能够将复杂的临床高维数据集通过计算简化为二维的可视化散点图。通过对散点图上重叠数据点的解析,快速地将疑难病例甄别出来,表明了数据分析模型的可靠性;研究结果表明超敏c反应蛋白可能在自身免疫性疾病的发生发展中具有启动者的作用;简化的检查项目组合也可以取得具有临床诊断价值的结果,在一定程度上节约了医疗资源。 展开更多
关键词 t-SNE(t-distributed STOCHASTIC NEIGHBOR Embedding) 自身免疫性疾病 数据分析 超敏C反应蛋白
下载PDF
High-impedance Fault Section Location for Distribution Networks Based on t-distributed Stochastic Neighbor Embedding and Variable Mode Decomposition
5
作者 Zhihua Yin Yuping Zheng +3 位作者 Zhinong Wei Guoqiang Sun Sheng Chen Haixiang Zang 《Journal of Modern Power Systems and Clean Energy》 SCIE EI 2024年第5期1495-1505,共11页
When high-impedance faults(HIFs)occur in resonant grounded distribution networks,the current that flows is extremely weak,and the noise interference caused by the distribution network operation and the sampling error ... When high-impedance faults(HIFs)occur in resonant grounded distribution networks,the current that flows is extremely weak,and the noise interference caused by the distribution network operation and the sampling error of the measurement devices further masks the fault characteristics.Consequently,locating a fault section with high sensitivity is difficult.Unlike existing technologies,this study presents a novel fault feature identification framework that addresses this issue.The framework includes three key steps:(1)utilizing the variable mode decomposition(VMD)method to denoise the fault transient zero-sequence current(TZSC);(2)employing a manifold learning algorithm based on t-distributed stochastic neighbor embedding(t-SNE)to further reduce the redundant information of the TZSC after denoising and to visualize fault information in high-dimensional 2D space;and(3)classifying the signal of each measurement point based on the fuzzy clustering method and combining the network topology structure to determine the fault section location.Numerical simulations and field testing confirm that the proposed method accurately detects the fault location,even under the influence of strong noise interference. 展开更多
关键词 High-impedance fault noise interference fault section location t-distributed stochastic neighbor embedding(t-SNE) transient zero-sequence current
原文传递
Reliability analysis method for slope stability based on sample weight
6
作者 Zhi-gang YANG Tong-chun LI Miao-lin DAI 《Water Science and Engineering》 EI CAS 2009年第3期78-86,共9页
The single safety factor criteria for slope stability evaluation, derived from the rigid limit equilibrium method or finite element method (FEM), may not include some important information, especially for steep slop... The single safety factor criteria for slope stability evaluation, derived from the rigid limit equilibrium method or finite element method (FEM), may not include some important information, especially for steep slopes with complex geological conditions. This paper presents a new reliability method that uses sample weight analysis. Based on the distribution characteristics of random variables, the minimal sample size of every random variable is extracted according to a small sample t-distribution under a certain expected value, and the weight coefficient of each extracted sample is considered to be its contribution to the random variables. Then, the weight coefficients of the random sample combinations are determined using the Bayes formula, and different sample combinations are taken as the input for slope stability analysis. According to one-to-one mapping between the input sample combination and the output safety coefficient, the reliability index of slope stability can be obtained with the multiplication principle. Slope stability analysis of the left bank of the Baihetan Project is used as an example, and the analysis results show that the present method is reasonable and practicable for the reliability analysis of steep slopes with complex geological conditions. 展开更多
关键词 reliability analysis slope stability sample weight coefficient t-distribution Bayes formula
下载PDF
ESTIMATING THE NONCENTRALITY PARAMETER OF A t-DISTRIBUTION
7
作者 Andrew L.Rukhin 《Systems Science and Mathematical Sciences》 SCIE EI CSCD 1992年第1期1-8,共8页
Inadmissibility of a traditional class of noncentrality parameter esti-mators under quadratic loss is established.The result is heuristically motivatedby the form of generalized Bayes estimators and is proved via unbi... Inadmissibility of a traditional class of noncentrality parameter esti-mators under quadratic loss is established.The result is heuristically motivatedby the form of generalized Bayes estimators and is proved via unbiased estimatorsof the risk function and a solution to an integro-differential inequality. 展开更多
关键词 Point estimation NORMAL parameters noncentral t-distribution QUADRATIC ADMISSIBILITY generalized BAYES ESTIMATORS Ricatti differential equation
原文传递
DSP-TMM:A Robust Cluster Analysis Method Based on Diversity Self-Paced T-Mixture Model
8
作者 Limin Pan Xiaonan Qin Senlin Luo 《Journal of Beijing Institute of Technology》 EI CAS 2020年第4期531-543,共13页
In order to implement the robust cluster analysis,solve the problem that the outliers in the data will have a serious disturbance to the probability density parameter estimation,and therefore affect the accuracy of cl... In order to implement the robust cluster analysis,solve the problem that the outliers in the data will have a serious disturbance to the probability density parameter estimation,and therefore affect the accuracy of clustering,a robust cluster analysis method is proposed which is based on the diversity self-paced t-mixture model.This model firstly adopts the t-distribution as the submodel which tail is easily controllable.On this basis,it utilizes the entropy penalty expectation conditional maximal algorithm as a pre-clustering step to estimate the initial parameters.After that,this model introduces l2,1-norm as a self-paced regularization term and developes a new ECM optimization algorithm,in order to select high confidence samples from each component in training.Finally,experimental results on several real-world datasets in different noise environments show that the diversity self-paced t-mixture model outperforms the state-of-the-art clustering methods.It provides significant guidance for the construction of the robust mixture distribution model. 展开更多
关键词 cluster analysis Gaussian mixture model t-distribution mixture model self-paced learning INITIALIZATION
下载PDF
Using Excel to Explore the Effects of Assumption Violations on One-Way Analysis of Variance (ANOVA) Statistical Procedures
9
作者 William Laverty Ivan Kelly 《Open Journal of Statistics》 2019年第4期458-469,共12页
To understand any statistical tool requires not only an understanding of the relevant computational procedures but also an awareness of the assumptions upon which the procedures are based, and the effects of violation... To understand any statistical tool requires not only an understanding of the relevant computational procedures but also an awareness of the assumptions upon which the procedures are based, and the effects of violations of these assumptions. In our earlier articles (Laverty, Miket, & Kelly [1]) and (Laverty & Kelly, [2] [3]) we used Microsoft Excel to simulate both a Hidden Markov model and heteroskedastic models showing different realizations of these models and the performance of the techniques for identifying the underlying hidden states using simulated data. The advantage of using Excel is that the simulations are regenerated when the spreadsheet is recalculated allowing the user to observe the performance of the statistical technique under different realizations of the data. In this article we will show how to use Excel to generate data from a one-way ANOVA (Analysis of Variance) model and how the statistical methods behave both when the fundamental assumptions of the model hold and when these assumptions are violated. The purpose of this article is to provide tools for individuals to gain an intuitive understanding of these violations using this readily available program. 展开更多
关键词 EXCEL One-Way ANOVA ASSUMPTION VIOLATIONS t-distribution CAUCHY Distribution
下载PDF
Test of Safety under Unequal Variances in Toxicological Studies
10
作者 郑术蓉 马文卿 《Northeastern Mathematical Journal》 CSCD 2003年第2期111-114,共4页
关键词 safety assessment student's t-distribution POWER sample size
下载PDF
Effective Truncation of a Student’s <i>t</i>-Distribution by Truncation of the Chi Distribution in a Chi-Normal Mixture
11
作者 Daniel T. Cassidy 《Open Journal of Statistics》 2012年第5期519-525,共7页
A Student’s t-distribution is obtained from a weighted average over the standard deviation of a normal distribution, σ, when 1/σ is distributed as chi. Left truncation at q of the chi distribution in the mixing int... A Student’s t-distribution is obtained from a weighted average over the standard deviation of a normal distribution, σ, when 1/σ is distributed as chi. Left truncation at q of the chi distribution in the mixing integral leads to an effectively truncated Student’s t-distribution with tails that decay as exp (-q2t2). The effect of truncation of the chi distribution in a chi-normal mixture is investigated and expressions for the pdf, the variance, and the kurtosis of the t-like distribution that arises from the mixture of a left-truncated chi and a normal distribution are given for selected degrees of freedom 5. This work has value in pricing financial assets, in understanding the Student’s t--distribution, in statistical inference, and in analysis of data. 展开更多
关键词 Asset Pricing Student’s t-distribution Cauchy TRUNCATION Moments Kurtosis
下载PDF
THE LIMITING DISTRIBUTIONS OF SOME SUBCLASSES OF THE GENERALIZED NON-CENTRAL t-DISTRIBUTION
12
作者 方开泰 袁克海 《Acta Mathematicae Applicatae Sinica》 SCIE CSCD 1993年第1期71-81,共11页
A general theorem on the limiting distribution of the generalized t-distribution is obtained,many applications of this theorem to some subclasses of elliptically contoured distributions in cluding multivariate normal ... A general theorem on the limiting distribution of the generalized t-distribution is obtained,many applications of this theorem to some subclasses of elliptically contoured distributions in cluding multivariate normal and multivariate t distributions are discussed.Further,their limiting distributions by density function are derived. 展开更多
关键词 exp THE LIMITING DISTRIBUTIONS OF SOME SUBCLASSES OF THE GENERALIZED NON-CENTRAL t-distribution
原文传递
Cryptographic Lightweight Encryption Algorithm with Dimensionality Reduction in Edge Computing
13
作者 D.Jerusha T.Jaya 《Computer Systems Science & Engineering》 SCIE EI 2022年第9期1121-1132,共12页
Edge Computing is one of the radically evolving systems through generations as it is able to effectively meet the data saving standards of consumers,providers and the workers. Requisition for Edge Computing based ite... Edge Computing is one of the radically evolving systems through generations as it is able to effectively meet the data saving standards of consumers,providers and the workers. Requisition for Edge Computing based items havebeen increasing tremendously. Apart from the advantages it holds, there remainlots of objections and restrictions, which hinders it from accomplishing the needof consumers all around the world. Some of the limitations are constraints oncomputing and hardware, functions and accessibility, remote administration andconnectivity. There is also a backlog in security due to its inability to create a trustbetween devices involved in encryption and decryption. This is because securityof data greatly depends upon faster encryption and decryption in order to transferit. In addition, its devices are considerably exposed to side channel attacks,including Power Analysis attacks that are capable of overturning the process.Constrained space and the ability of it is one of the most challenging tasks. Toprevail over from this issue we are proposing a Cryptographic LightweightEncryption Algorithm with Dimensionality Reduction in Edge Computing. Thet-Distributed Stochastic Neighbor Embedding is one of the efficient dimensionality reduction technique that greatly decreases the size of the non-linear data. Thethree dimensional image data obtained from the system, which are connected withit, are dimensionally reduced, and then lightweight encryption algorithm isemployed. Hence, the security backlog can be solved effectively using thismethod. 展开更多
关键词 Edge computing(e.g) dimensionality reduction(dr) t-distributed stochastic neighbor embedding(t-sne) principle component analysis(pca)
下载PDF
Student’s t Increments
14
作者 Daniel T. Cassidy 《Open Journal of Statistics》 2016年第1期156-171,共16页
Some moments and limiting properties of independent Student’s t increments are studied. Inde-pendent Student’s t increments are independent draws from not-truncated, truncated, and effectively truncated Student’s t... Some moments and limiting properties of independent Student’s t increments are studied. Inde-pendent Student’s t increments are independent draws from not-truncated, truncated, and effectively truncated Student’s t-distributions with shape parameters and can be used to create random walks. It is found that sample paths created from truncated and effectively truncated Student’s t-distributions are continuous. Sample paths for Student’s t-distributions are also continuous. Student’s  t increments should thus be useful in construction of stochastic processes and as noise driving terms in Langevin equations. 展开更多
关键词 Student’s t-distribution TRUNCATED Effectively Truncated Cauchy Distribution Random Walk Sample Paths CONTINUITY
下载PDF
Amplitude variation with offset and azimuth inversion to predict and evaluate coal seam fracture parameters
15
作者 Haibo WU Shujie ZHU +3 位作者 Qinjie LIU Shouhua DONG Yanhui HUANG Pingsong ZHANG 《Frontiers of Earth Science》 SCIE CSCD 2023年第2期505-513,共9页
Amplitude variation with offset and azimuth(AVOA)inversion is a mainstream method for predicting and evaluating fracture parameters of conventional oil and gas reservoirs.However,its application to coal seams is limit... Amplitude variation with offset and azimuth(AVOA)inversion is a mainstream method for predicting and evaluating fracture parameters of conventional oil and gas reservoirs.However,its application to coal seams is limited because of the specificity of the equivalent media model for coal—also,the traditional seismic acquisition system employed in coal fields falls within a narrow azimuth.In this study,we initially derived a P‒P wave reflection coefficient approximation formula for coal seams,which is directly expressed in terms of fracture parameters using the Schoenberg linear-slide model and Hudson model.We analyzed the P‒P wave reflection coefficient’s response to the fracture parameters using a two-layer forward model.Accordingly,we designed a twostep inversion workflow for AVOA inversion of the fracture parameters.Thereafter,high-density wide-azimuth pre-stack 3D seismic data were utilized for inverting the fracture density and strike of the target coal seam.The inversion accuracy was constrained by Student’s tdistribution testing.The analysis and validation of the inversion results revealed that the relative fracture density corresponds to fault locations,with the strike of the fractures and faults mainly at 0°.Therefore,the AVOA inversion method and technical workflow proposed here can be used to efficiently predict and evaluate fracture parameters of coal seams. 展开更多
关键词 equivalent media model fracture density and strike AZIMUTH Student’s t-distribution
原文传递
A TEST FOR SPHERICITY OF ERRORS IN LINEAR MODELS 被引量:1
16
作者 WANG Songgui(Institute of Statistics, Beijing Polytechnic University, Beijing 100022, China)(Institute of Applied of Mathematics, Academia Sinica, Beijing 100080, China)YANG Zhenhai(Institute of Statistics, Beijing Polytechnic University, Beijing 100022, 《Systems Science and Mathematical Sciences》 SCIE EI CSCD 1997年第3期202-207,共6页
For a general linear model, spherical distributions are often considered whenerrors do not have normal distribution. Several authors[1-3] studied the least squaresand James-Stein estimations for a linear model whose e... For a general linear model, spherical distributions are often considered whenerrors do not have normal distribution. Several authors[1-3] studied the least squaresand James-Stein estimations for a linear model whose errors follow multivariate t or moregeneral spherical distributions. In this paper the test problem for sphericity of errors isconsidered. We propose an exact test for the sphericity by using the conditional probabilityintegral transformation and another transformation. As an important special case, thecorresponding test statistics for multivariate t distribution are obtained. 展开更多
关键词 Spherical DISTRIBUTION MULTIVARIATE t-distribution conditional probability integral transformation Pearson’s product TEST MULTIVARIATE CAUCHY distribution.
原文传递
Large-Scale Estimation of Distribution Algorithms with Adaptive Heavy Tailed Random Pro jection Ensembles
17
作者 Momodou L.Sanyang Ata Kabán 《Journal of Computer Science & Technology》 SCIE EI CSCD 2019年第6期1241-1257,共17页
We present new variants of Estimation of Distribution Algorithms (EDA) for large-scale continuous optimisation that extend and enhance a recently proposed random projection (RP) ensemble based approach. The main novel... We present new variants of Estimation of Distribution Algorithms (EDA) for large-scale continuous optimisation that extend and enhance a recently proposed random projection (RP) ensemble based approach. The main novelty here is to depart from the theory of RPs that require (sub-)Gaussian random matrices for norm-preservation, and instead for the purposes of high-dimensional search we propose to employ random matrices with independent and identically distributed entries drawn from a t-distribution. We analytically show that the implicitly resulting high-dimensional covariance of the search distribution is enlarged as a result. Moreover, the extent of this enlargement is controlled by a single parameter, the degree of freedom. For this reason, in the context of optimisation, such heavy tailed random matrices turn out to be preferable over the previously employed (sub-)Gaussians. Based on this observation, we then propose novel covariance adaptation schemes that are able to adapt the degree of freedom parameter during the search, and give rise to a flexible approach to balance exploration versus exploitation. We perform a thorough experimental study on high-dimensional benchmark functions, and provide statistical analyses that demonstrate the state-of-the-art performance of our approach when compared with existing alternatives in problems with 1000 search variables. 展开更多
关键词 covariance adaptation estimation of distribution algorithm random projection ensemble t-distribution
原文传递
Light spectrum preference of Nile Tilapia (Oreochromis niloticus) under different hunger levels
18
作者 Guang Jin Jian Zhao +5 位作者 Yadong Zhang Gang Liu Dezhao Liu Songming Zhu Yufang Shao Zhangying Ye 《International Journal of Agricultural and Biological Engineering》 SCIE EI CAS 2019年第5期51-57,共7页
In order to improve the light welfare of Nile tilapia in aquaculture,the influence of hunger level on light spectrum preference of Nile tilapia was explored in this study.The whole experiment was based on the emptying... In order to improve the light welfare of Nile tilapia in aquaculture,the influence of hunger level on light spectrum preference of Nile tilapia was explored in this study.The whole experiment was based on the emptying of the gastrointestinal contents,and carried out under the controlled laboratory conditions.The light spectrum preference was assessed by counting the head location of fish in each experimental tank,which containing seven compartments(i.e.,red,blue,white,yellow,black,green and public area).t-Distributed Stochastic Neighbor Embedding(t-SNE)was adopted to visualize the hunger level-based dynamic preference on light spectrum in two-dimensional space.According to the clustering results,significant differences in light spectrum preferences of Nile tilapia,under the different hunger levels,were indicated.In addition,the average visit frequency in green compartment was significantly lower than that in other color compartments throughout the whole experiment,and the total visit frequency in red compartment was relatively higher during the whole experiment. 展开更多
关键词 light welfare Nile tilapia hunger level light spectrum preference t-distributed Stochastic Neighbor Embedding
原文传递
Information Divergence and the Generalized Normal Distribution:A Study on Symmetricity
19
作者 Thomas L.Toulias Christos P.Kitsos 《Communications in Mathematics and Statistics》 SCIE 2021年第4期439-465,共27页
This paper investigates and discusses the use of information divergence,through the widely used Kullback–Leibler(KL)divergence,under the multivariate(generalized)γ-order normal distribution(γ-GND).The behavior of t... This paper investigates and discusses the use of information divergence,through the widely used Kullback–Leibler(KL)divergence,under the multivariate(generalized)γ-order normal distribution(γ-GND).The behavior of the KL divergence,as far as its symmetricity is concerned,is studied by calculating the divergence of γ-GND over the Student’s multivariate t-distribution and vice versa.Certain special cases are also given and discussed.Furthermore,three symmetrized forms of the KL divergence,i.e.,the Jeffreys distance,the geometric-KL as well as the harmonic-KL distances,are computed between two members of the γ-GND family,while the corresponding differences between those information distances are also discussed. 展开更多
关键词 Kullback-Leibler divergence Jeffreys distance Resistor-average distance Multivariateγ-order normal distribution Multivariate Student’s t-distribution Multivariate Laplace distribution
原文传递
Learning Gaussian mixture with automatic model selection:A comparative study on three Bayesian related approaches
20
作者 Lei SHI Shikui TU Lei XU 《Frontiers of Electrical and Electronic Engineering in China》 CSCD 2011年第2期215-244,共30页
Three Bayesian related approaches,namely,variational Bayesian(VB),minimum message length(MML)and Bayesian Ying-Yang(BYY)harmony learning,have been applied to automatically determining an appropriate number of componen... Three Bayesian related approaches,namely,variational Bayesian(VB),minimum message length(MML)and Bayesian Ying-Yang(BYY)harmony learning,have been applied to automatically determining an appropriate number of components during learning Gaussian mixture model(GMM).This paper aims to provide a comparative investigation on these approaches with not only a Jeffreys prior but also a conjugate Dirichlet-Normal-Wishart(DNW)prior on GMM.In addition to adopting the existing algorithms either directly or with some modifications,the algorithm for VB with Jeffreys prior and the algorithm for BYY with DNW prior are developed in this paper to fill the missing gap.The performances of automatic model selection are evaluated through extensive experiments,with several empirical findings:1)Considering priors merely on the mixing weights,each of three approaches makes biased mistakes,while considering priors on all the parameters of GMM makes each approach reduce its bias and also improve its performance.2)As Jeffreys prior is replaced by the DNW prior,all the three approaches improve their performances.Moreover,Jeffreys prior makes MML slightly better than VB,while the DNW prior makes VB better than MML.3)As the hyperparameters of DNW prior are further optimized by each of its own learning principle,BYY improves its performances while VB and MML deteriorate their performances when there are too many free hyper-parameters.Actually,VB and MML lack a good guide for optimizing the hyper-parameters of DNW prior.4)BYY considerably outperforms both VB and MML for any type of priors and whether hyper-parameters are optimized.Being different from VB and MML that rely on appropriate priors to perform model selection,BYY does not highly depend on the type of priors.It has model selection ability even without priors and performs already very well with Jeffreys prior,and incrementally improves as Jeffreys prior is replaced by the DNW prior.Finally,all algorithms are applied on the Berkeley segmentation database of real world images.Again,BYY considerably outperforms both VB and MML,especially in detecting the objects of interest from a confusing background. 展开更多
关键词 Bayesian Ying-Yang(BYY)harmony learning variational Bayesian(VB) minimum message length(MML) empirical comparison Gaussian mixture model(GMM) automatic model selection Jeffreys prior DIRICHLET joint Normal-Wishart(NW) conjugate distributions marginalized student’s t-distribution
原文传递
上一页 1 下一页 到第
使用帮助 返回顶部