We propose a new method for learning the parameters of a Bayesian network with qualitative influences. The proposed method aims to remove unwanted (context-specific) independencies that are created by the order-constrained maximum likelihood (OCML) estimator. This is achieved by averaging the OCML estimator with the fitted probabilities of a first-order logistic regression model. We show experi...

A simultaneous multi-frame super-resolution video reconstruction procedure, utilizing spatio-temporal smoothness constraints and motion estimator confidence parameters is proposed. The ill-posed inverse problem of reconstructing super-resolved imagery from the low resolution, degraded observations is formulated as a statistical inference problem and a Bayesian, maximum a-posteriori (MAP) approa...

In this paper we propose a generalised iterative algorithm for calculating variational Bayesian estimates for a normal mixture model and we investigate its convergence properties. It is shown theoretically that the variational Bayes estimator converges locally to the maximum likelihood estimator at the rate of O(1/n) in the large sample limit. We also demonstrate by numerical experiments that t...

We employ Lasso shrinkage within the context of sufficient dimension reduction to obtain a shrinkage sliced inverse regression estimator, which provides easier interpretations and better prediction accuracy without assuming a parametric model. The shrinkage sliced inverse regression approach can be employed for both single-index and multiple-index models. Simulation studies suggest that the new...

Estimation on a noisy signal observed by a nonlinear sensor taking the form of a threshold quantizer is considered. The optimal Bayesian estimator with minimal error is derived in this nonlinear setting. The existence of conditions where the performance of this estimator can be improved by raising the level of noise is established, both theoretically and numerically. These results constitute a ...

The integrated likelihood (also called the marginal likelihood or the normalizing constant) is a central quantity in Bayesian model selection and model averaging. It is defined as the integral over the parameter space of the likelihood times the prior density. The Bayes factor for model comparison and Bayesian testing is a ratio of integrated likelihoods, and the model weights in Bayesian model...

The key quantity needed for Bayesian hypothesis testing and model selection is the marginal likelihood for a model, also known as the integrated likelihood, or the marginal probability of the data. In this paper we describe a way to use posterior simulation output to estimate marginal likelihoods. vVe describe the basic LaplaceMetropolis estimator for models without random effects. For models w...

This paper develops Bayesian penalized spline predictive (BPSP) estimator of finite population proportion for probability-proportional-to-size samples. This new method allows the probabilities of inclusion to be directly incorporated into the estimation of population proportion, using a probit regression of the binary outcome on the penalized spline of the inclusion probabilities. The posterior...

We use the model resolution matrix to analytically derive an optimal Bayesian estimator for multiparameter inverse problems that simultaneously minimizes inter-parameter cross talk and the total reconstruction error. Application of this estimator to time-domain diffuse fluorescence imaging shows that the optimal estimator for lifetime multiplexing is identical to a previously developed asymptot...

Massive MIMO systems that for a cellular network, the channel from user equipment to a base station is composed of few grouped paths in space. With a very large antenna array, signals can be observed under extremely sharp regions in space. In the FDD mode, each BS sends a downlink training matrix to its served UEs which estimates the desired channel based on the downlink measurements and feeds ...