Bayesian mixture models in a longitudinal setting for analysing sheep CAT scan images

CAT scanning is used in longitudinal animal science experiments to assess possible changes to carcase composition induced by treatment over given time periods. A hierarchical Bayesian mixture model can be used to analyse the CAT scan data in terms of the proportion of each tissue type present in a scan. In this paper we present an extension to the hierarchical Bayesian mixture model in which estimated parameters from neighbouring CAT scans can be incorporated into the current model. These models are demonstrated using two examples.     

Bayesian hybrid generative discriminative learning based on finite Liouville mixture models

Recently hybrid generative discriminative approaches have emerged as an efficient knowledge representation and data classification engine. However, little attention has been devoted to the modeling and classification of non-Gaussian and especially proportional vectors. Our main goal, in this paper, is to discover the true structure of this kind of data by building probabilistic kernels from generative mixture models based on Liouville family, from which we develop the Beta-Liouville distribution, and which includes the well-known Dirichlet as a special case. The Beta-Liouville has a more general covariance structure than the Dirichlet which makes it more practical and useful. Our learning technique is based on a principled purely Bayesian approach which resulted models are used to generate support vector machine (SVM) probabilistic kernels based on information divergence. In particular, we show the existence of closed-form expressions of the Kullback-Leibler and Rényi divergences between two Beta-Liouville distributions and then between two Dirichlet distributions as a special case. Through extensive simulations and a number of experiments involving synthetic data, visual scenes and texture images classification, we demonstrate the effectiveness of the proposed approaches.     

A version of the Swendsen-Wang algorithm for restoration of images degraded by Poisson noise

An algorithm for restoration of images degraded by Poisson noise is proposed. The algorithm belongs to the family of Markov chain Monte Carlo methods with auxiliary variables. We explicitly use the fact that medical images consist of finitely many, often relatively few, grey-levels. The continuous scale of grey-levels is discretized in an adaptive way, so that a straightforward application of the Swendsen-Wang (Phys. Rev. Lett. 58 (1987) 86) algorithm becomes possible. Partial decoupling method due to Higdon (J. Am. Statist. Assoc. 93 (1998) 442, 585) is also incorporated into the algorithm. Simulation results suggest that the algorithm is reliable and efficient.     

A novel hybrid learning algorithm for full Bayesian approach of artificial neural networks

The Bayesian neural networks are useful tools to estimate the functional structure in the nonlinear systems. However, they suffer from some complicated problems such as controlling the model complexity, the training time, the efficient parameter estimation, the random walk, and the stuck in the local optima in the high-dimensional parameter cases. In this paper, to alleviate these mentioned problems, a novel hybrid Bayesian learning procedure is proposed. This approach is based on the full Bayesian learning, and integrates Markov chain Monte Carlo procedures with genetic algorithms and the fuzzy membership functions. In the application sections, to examine the performance of proposed approach, nonlinear time series and regression analysis are handled separately, and it is compared with the traditional training techniques in terms of their estimation and prediction abilities.     

Some extensions of score matching

Many probabilistic models are only defined up to a normalization constant. This makes maximum likelihood estimation of the model parameters very difficult. Typically, one then has to resort to Markov Chain Monte Carlo methods, or approximations of the normalization constant. Previously, a method called score matching was proposed for computationally efficient yet (locally) consistent estimation of such models. The basic form of score matching is valid, however, only for models which define a differentiable probability density function over Rⁿ. Therefore, some extensions of the framework are proposed. First, a related method for binary variables is proposed. Second, it is shown how to estimate non-normalized models defined in the non-negative real domain, i.e. . As a further result, it is shown that the score matching estimator can be obtained in closed form for some exponential families.     

Comparison of estimates using record statistics from Weibull model: Bayesian and non-Bayesian approaches

This paper develops a Bayesian analysis in the context of record statistics values from the two-parameter Weibull distribution. The ML and the Bayes estimates based on record values are derived for the two unknown parameters and some survival time parameters e.g. reliability and hazard functions. The Bayes estimates are obtained based on a conjugate prior for the scale parameter and a discrete prior for the shape parameter of this model. This is done with respect to both symmetric loss function (squared error loss), and asymmetric loss function (linear-exponential (LINEX)) loss function. The maximum likelihood and the different Bayes estimates are compared via a Monte Carlo simulation study. A practical example consisting of real record values using the data from an accelerated test on insulating fluid reported by Nelson was used for illustration and comparison. Finally, Bayesian predictive density function, which is necessary to obtain bounds for predictive interval of future record is derived and discussed using a numerical example. The results may be of interest in a situation where only record values are stored.     

A Bayesian analysis of an agricultural field trial with three spatial dimensions

Modern technology now has the ability to generate large datasets over space and time. Such data typically exhibit high autocorrelations over all dimensions. The field trial data motivating the methods of this paper were collected to examine the behaviour of traditional cropping and to determine a cropping system which could maximise water use for grain production while minimising leakage below the crop root zone. They consist of moisture measurements made at 15 depths across 3 rows and 18 columns, in the lattice framework of an agricultural field.Bayesian conditional autoregressive (CAR) models are used to account for local site correlations. Conditional autoregressive models have not been widely used in analyses of agricultural data. This paper serves to illustrate the usefulness of these models in this field, along with the ease of implementation in WinBUGS, a freely available software package.The innovation is the fitting of separate conditional autoregressive models for each depth layer, the ‘layered CAR model’, while simultaneously estimating depth profile functions for each site treatment. Modelling interest also lies in how best to model the treatment effect depth profiles, and in the choice of neighbourhood structure for the spatial autocorrelation model. The favoured model fitted the treatment effects as splines over depth, and treated depth, the basis for the regression model, as measured with error, while fitting CAR neighbourhood models by depth layer. It is hierarchical, with separate conditional autoregressive spatial variance components at each depth, and the fixed terms which involve an errors-in-measurement model treat depth errors as interval-censored measurement error. The Bayesian framework permits transparent specification and easy comparison of the various complex models compared.