检索结果-内蒙古大学图书馆

Enhanced-SAGE algorithm for use in distributed-source environments

ELECTRONICS LETTERS 2003年第8期39卷 697-698页

作者： Tan, CM Beach, MA Nix, AR Univ Bristol Ctr Commun Res Bristol BSB 1UB Avon England

An enhanced version of the space-alternating generalised expectation maximisation (SAGE) algorithm based on distributed-source modelling is proposed. This new algorithm shows an improved performance compared to that of the classical SAGE algorithm when applied to a distributed-source environment, especially when the successive interference cancellation (SIC) technique is employed within the SAGE algorithm.

关键词： Mathematical models expectation-maximization algorithms

来源：评论

学校读者我要写书评

暂无评论

ESTIMATORS OF SEVERITY FACTORS IN A MULTIPLICATIVE POISSON MODEL

引用

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION 1972年第337期67卷 192-&页

作者： ZACKS, S ZIMMER, WJ Professor of Mathematics and Statistics Case Western Reserve University Cleveland Ohio 44106. Associate Professor of mathematics and statistics University of New Mexico Albuquerque N. M. 87100

Let Yij(i=1, …, lii=1, …, J) be independent Poisson random variables, with expectations λ= αiβjiwhere Σαi= l. Such a model is called σ multiplicative Poisson; its factor parameters αi, are called severity factors, and β, are called intensity factors. The article concentrates on the problem of estimating the severity factors αi. Two types of estimators of αiare derived, Bayes and least-squares-maximum-likelihood. The expectations and mean square errors of these estimators are given and their relative efficiency is tabulated. The intensity factors β; are estimated from Ti= Σ Ii=1yij, which are independent and have Poisson distributions with expectations θj= lβj, according to the common procedures.

关键词： RANDOM variables ESTIMATION theory MATHEMATICAL models MATHEMATICAL statistics STATISTICS MULTIVARIATE analysis POISSON distribution POISSON processes expectation-maximization algorithms LEAST squares

来源：评论

学校读者我要写书评

暂无评论

Identification of a Markovian system with observations corrupted by a fractional Brownian motion

引用

STATISTICS & PROBABILITY LETTERS 2009年第7期79卷 965-968页

作者： Mandrekar, V. Naik-Nimbalkar, U. V. Univ Poona Dept Stat Pune 411007 Maharashtra India Michigan State Univ Dept Stat & Probabil E Lansing MI 48824 USA

We examine the continuous time analogue of the work of [Shumway, R.H., Stoffer, D.S., 1982. An approach to time series smoothing and forecasting using EM algorithm. J. Time Ser. 3, 253-264] for state space models when the noise in the observation process is a fractional Brownian motion. We Study the estimation problem for the parameter of the system process. (c) 2008 Elsevier B.V. All rights reserved.

关键词： MARKOV processes BROWNIAN motion processes expectation-maximization algorithms CONTINUOUS functions ERROR analysis (Mathematics) TIME series analysis PARAMETER estimation

来源：评论

学校读者我要写书评

暂无评论

MIXED MODELS FOR BINOMIAL DATA WITH AN APPLICATION TO LAMB MORTALITY

引用

JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES C-APPLIED STATISTICS 1988年第2期37卷 196-204页

作者： IM, S GIANOLA, D INRA F-31300 TOULOUSE FRANCE UNIV ILLINOIS URBANA IL 61801 USA

SUMMARY The simplex method, a derivative-free function maximisation algorithm, is used as an alternative to the EM algorithm in computing maximum likelihood estimates in mixed probit and logit models with binomial data. The models are used to estimate heritability and to predict sire effects when analysing a lamb mortality data set. [ABSTRACT FROM AUTHOR]

关键词： BINOMIAL distribution LAMBS expectation-maximization algorithms MORTALITY

来源：评论

学校读者我要写书评

暂无评论

Probabilistic partitioning methods to find significant patterns in ChIP-Seq data

引用

BIOINFORMATICS 2014年第17期30卷 2406-2413页

作者： Nair, Nishanth Ulhas Kumar, Sunil Moret, Bernard M. E. Bucher, Philipp Ecole Polytech Fed Lausanne Sch Comp & Commun Sci Lab Computat Biol & Bioinformat CH-1015 Lausanne Switzerland Ecole Polytech Fed Lausanne Sch Life Sci Swiss Inst Expt Canc Res ISREC CH-1015 Lausanne Switzerland Swiss Inst Bioinformat CH-1015 Lausanne Switzerland

Motivation: We have witnessed an enormous increase in ChIP-Seq data for histone modifications in the past few years. Discovering significant patterns in these data is an important problem for understanding biological mechanisms. Results: We propose probabilistic partitioning methods to discover significant patterns in ChIP-Seq data. Our methods take into account signal magnitude, shape, strand orientation and shifts. We compare our methods with some current methods and demonstrate significant improvements, especially with sparse data. Besides pattern discovery and classification, probabilistic partitioning can serve other purposes in ChIP-Seq data analysis. Specifically, we exemplify its merits in the context of peak finding and partitioning of nucleosome positioning patterns in human promoters.

关键词： SEQUENCE alignment (Bioinformatics) PROBABILITY theory PARALLEL algorithms IMMUNOPRECIPITATION HIGH throughput screening (Drug development) EUCLIDEAN distance expectation-maximization algorithms

来源：评论

学校读者我要写书评

暂无评论

Conditional Mixture Model for Correlated Neuronal Spikes

引用

NEURAL COMPUTATION 2010年第7期22卷 1718-1736页

作者： Amari, Shun-ichi RIKEN Brain Sci Inst Lab Math Neurosci Wako Saitama 3510198 Japan

Analysis of correlated spike trains is a hot topic of research in computational neuroscience. A general model of probability distributions for spikes includes too many parameters to be of use in analyzing real data. Instead, we need a simple but powerful generative model for correlated spikes. We developed a class of conditional mixture models that includes a number of existing models and analyzed its capabilities and limitations. We apply the model to dynamical aspects of neuron pools. When Hebbian cell assemblies coexist in a pool of neurons, the condition is specified by these assemblies such that the probability distribution of spikes is a mixture of those of the component assemblies. The probabilities of activation of the Hebbian assemblies change dynamically. We used this model as a basis for a competitive model governing the states of assemblies.

关键词： NEUROSCIENCES POISSON processes PROBABILITY theory MATHEMATICAL statistics NEURONS expectation-maximization algorithms DATA analysis

来源：评论

学校读者我要写书评

暂无评论

Nonparametric estimation of renewal processes from count data (vol 31, pg 191, 2003)

引用

CANADIAN JOURNAL OF STATISTICS-REVUE CANADIENNE DE STATISTIQUE 2003年第4期31卷 483-483页

作者： Guèdon, Y Cocozza-Thivent, C Univ Montpellier 2 INRA CNRS CIRADUnite Mixte Rech FR-34398 Montpellier 5 France Univ Marne la Vallee Lab Anal & Math Appl FR-77454 Marne La Vallee 2 France

The authors address the problem of estimating an inter-event distribution on the basis of count data. They derive a nonparametric maximum likelihood estimate of the inter-event distribution utilizing the EM algorithm both in the case of an ordinal renewal process and in the case of an equilibrium renewal process. In the latter case, the iterative estimation procedure follows the basic scheme proposed by Vardi for estimating the inter-event distribution on the basis of time-interval data; it combines the outputs of the E-step corresponding to the inter-event distribution and to the length-biased distribution. The authors also investigate a penalized likelihood approach to provide the proposed estimation procedures with regularization capabilities. They evaluate the practical estimation procedure using simulated count data and apply it to real count data representing the elongation of coffee-tree leafy axes.

关键词： expectation-maximization algorithms NONPARAMETRIC statistics MAXIMUM likelihood statistics PLANT growth ESTIMATION theory

来源：评论

学校读者我要写书评

暂无评论

Discrete- and Continuous-Time Probabilistic Models and algorithms for Inferring Neuronal UP and DOWN States

引用

NEURAL COMPUTATION 2009年第7期21卷 1797-1862页

作者： Chen, Zhe Vijayan, Sujith Barbieri, Riccardo Wilson, Matthew A. Brown, Emery N. Harvard Univ Massachusetts Gen Hosp Sch Med Neurosci Stat Res LabDept Anesthesia & Crit Care Boston MA 02114 USA MIT Harvard Mit Div Hlth Sci & Technol Cambridge MA 02139 USA Harvard Univ Program Neurosci Cambridge MA 02139 USA MIT Dept Biol Cambridge MA 02139 USA MIT Picower Inst Learning & Memory RIKEN MIT Neurosci Res Ctr Dept Brain & Cognit Sci Cambridge MA 02139 USA

UP and DOWN states, the periodic fluctuations between increased and decreased spiking activity of a neuronal population, are a fundamental feature of cortical circuits. Understanding UP-DOWN state dynamics is important for understanding how these circuits represent and transmit information in the brain. To date, limited work has been done on characterizing the stochastic properties of UP-DOWN state dynamics. We present a set of Markov and semi-Markov discrete-and continuous-time probability models for estimating UP and DOWN states from multiunit neural spiking activity. We model multiunit neural spiking activity as a stochastic point process, modulated by the hidden (UP and DOWN) states and the ensemble spiking history. We estimate jointly the hidden states and the model parameters by maximum likelihood using an expectation-maximization (EM) algorithm and a Monte Carlo EM algorithm that uses reversible-jump Markov chain Monte Carlo sampling in the E-step. We apply our models and algorithms in the analysis of both simulated multiunit spiking activity and actual multiunit spiking activity recorded from primary somatosensory cortex in a behaving rat during slow-wave sleep. Our approach provides a statistical characterization of UP-DOWN state dynamics that can serve as a basis for verifying and refining mechanistic descriptions of this process.

关键词： NEURAL circuitry PROBABILITY theory STOCHASTIC processes expectation-maximization algorithms SOMATOSENSORY evoked potentials CEREBRAL cortex

来源：评论

学校读者我要写书评

暂无评论

AN EVALUATION OF THE SIEVES FOR PRODUCING ESTIMATES OF RADIOACTIVITY DISTRIBUTIONS WITH THE EM ALGORITHM FOR PET

引用

IEEE TRANSACTIONS ON NUCLEAR SCIENCE 1986年第1期33卷 492-495页

作者： MILLER, MI SNYDER, DL MOORE, SM WASHINGTON UNIV INST BIOMED COMPST LOUISMO 63130

Reconstructions of spatial-distributions of radioactivity produced using maximum-likelihood estimation in positron-emission tomography exhibit noise-like artifacts in the form of sharp peaks and valleys located randomly throughout the image field. These become increasingly apparent with each stage of iteration when the expectation-maximization algorithm is used to produce the maximum-likelihood estimate numerically. In this paper, we present a preliminary evaluation of the use of Grenander's method of sieves to reduce these undesirable artifacts.

关键词： Positron emission tomography Image reconstruction Maximum likelihood estimation expectation-maximization algorithms Maximum likelihood detection Signal resolution Laboratories Signal to noise ratio Biomedical computing Single photon emission computed tomography

来源：评论

学校读者我要写书评

暂无评论

GP-Select: Accelerating EM Using Adaptive Subspace Preselection

引用

NEURAL COMPUTATION 2017年第8期29卷 2177-2202页

作者： Shelton, Jacquelyn A. Gasthaus, Jan Dai, Zhenwen Luecke, Joerg Gretton, Arthur Tech Univ Berlin D-10587 Berlin Germany UCL Gatsby Unit London W1T 4JG England Amazon Dev Ctr D-10178 Berlin Germany Univ Sheffield Sheffield S10 2TN S Yorkshire England Carl von Ossietzky Univ Oldenburg D-26129 Oldenburg Germany

We propose a nonparametric procedure to achieve fast inference in generative graphical models when the number of latent states is very large. The approach is based on iterative latent variable preselection, where we alternate between learning a selection function to reveal the relevant latent variables and using this to obtain a compact approximation of the posterior distribution for EM. This can make inference possible where the number of possible latent states is, for example, exponential in the number of latent variables, whereas an exact approach would be computationally infeasible. We learn the selection function entirely from the observed data and current expectation-maximization state via gaussian process regression. This is in contrast to earlier approaches, where selection functions were manually designed for each problem setting. We show that our approach performs as well as these bespoke selection functions on a wide variety of inference problems. In particular, for the challenging case of a hierarchical model for object localization with occlusion, we achieve results that match a customized state-of-the-art selection method at a far lower computational cost.

关键词： expectation-maximization algorithms PROBABILITY theory APPROXIMATION theory COMBINATORICS MATHEMATICS

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：