96 resultados para Expectation-conditional Maximization (ecm)

em University of Queensland eSpace - Australia


Relevância:

100.00% 100.00%

Publicador:

Resumo:

We consider a mixture model approach to the regression analysis of competing-risks data. Attention is focused on inference concerning the effects of factors on both the probability of occurrence and the hazard rate conditional on each of the failure types. These two quantities are specified in the mixture model using the logistic model and the proportional hazards model, respectively. We propose a semi-parametric mixture method to estimate the logistic and regression coefficients jointly, whereby the component-baseline hazard functions are completely unspecified. Estimation is based on maximum likelihood on the basis of the full likelihood, implemented via an expectation-conditional maximization (ECM) algorithm. Simulation studies are performed to compare the performance of the proposed semi-parametric method with a fully parametric mixture approach. The results show that when the component-baseline hazard is monotonic increasing, the semi-parametric and fully parametric mixture approaches are comparable for mildly and moderately censored samples. When the component-baseline hazard is not monotonic increasing, the semi-parametric method consistently provides less biased estimates than a fully parametric approach and is comparable in efficiency in the estimation of the parameters for all levels of censoring. The methods are illustrated using a real data set of prostate cancer patients treated with different dosages of the drug diethylstilbestrol. Copyright (C) 2003 John Wiley Sons, Ltd.

Relevância:

100.00% 100.00%

Publicador:

Resumo:

The expectation-maximization (EM) algorithm has been of considerable interest in recent years as the basis for various algorithms in application areas of neural networks such as pattern recognition. However, there exists some misconceptions concerning its application to neural networks. In this paper, we clarify these misconceptions and consider how the EM algorithm can be adopted to train multilayer perceptron (MLP) and mixture of experts (ME) networks in applications to multiclass classification. We identify some situations where the application of the EM algorithm to train MLP networks may be of limited value and discuss some ways of handling the difficulties. For ME networks, it is reported in the literature that networks trained by the EM algorithm using iteratively reweighted least squares (IRLS) algorithm in the inner loop of the M-step, often performed poorly in multiclass classification. However, we found that the convergence of the IRLS algorithm is stable and that the log likelihood is monotonic increasing when a learning rate smaller than one is adopted. Also, we propose the use of an expectation-conditional maximization (ECM) algorithm to train ME networks. Its performance is demonstrated to be superior to the IRLS algorithm on some simulated and real data sets.

Relevância:

30.00% 30.00%

Publicador:

Resumo:

Objective: Inpatient length of stay (LOS) is an important measure of hospital activity, health care resource consumption, and patient acuity. This research work aims at developing an incremental expectation maximization (EM) based learning approach on mixture of experts (ME) system for on-line prediction of LOS. The use of a batchmode learning process in most existing artificial neural networks to predict LOS is unrealistic, as the data become available over time and their pattern change dynamically. In contrast, an on-line process is capable of providing an output whenever a new datum becomes available. This on-the-spot information is therefore more useful and practical for making decisions, especially when one deals with a tremendous amount of data. Methods and material: The proposed approach is illustrated using a real example of gastroenteritis LOS data. The data set was extracted from a retrospective cohort study on all infants born in 1995-1997 and their subsequent admissions for gastroenteritis. The total number of admissions in this data set was n = 692. Linked hospitalization records of the cohort were retrieved retrospectively to derive the outcome measure, patient demographics, and associated co-morbidities information. A comparative study of the incremental learning and the batch-mode learning algorithms is considered. The performances of the learning algorithms are compared based on the mean absolute difference (MAD) between the predictions and the actual LOS, and the proportion of predictions with MAD < 1 day (Prop(MAD < 1)). The significance of the comparison is assessed through a regression analysis. Results: The incremental learning algorithm provides better on-line prediction of LOS when the system has gained sufficient training from more examples (MAD = 1.77 days and Prop(MAD < 1) = 54.3%), compared to that using the batch-mode learning. The regression analysis indicates a significant decrease of MAD (p-value = 0.063) and a significant (p-value = 0.044) increase of Prop(MAD

Relevância:

20.00% 20.00%

Publicador:

Resumo:

This article considers the conditions placed on the autonomous architectural history discipline often understood at stake in Manfredo Tafuri's 1968 book Teorie e storia dell'architettura.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

Carbon monoxide is the chief killer in fires. Dangerous levels of CO can occur when reacting combustion gases are quenched by heat transfer, or by mixing of the fire plume in a cooled under- or overventilated upper layer. In this paper, carbon monoxide predictions for enclosure fires are modeled by the conditional moment closure (CMC) method and are compared with laboratory data. The modeled fire situation is a buoyant, turbulent, diffusion flame burning under a hood. The fire plume entrains fresh air, and the postflame gases are cooled considerably under the hood by conduction and radiation, emulating conditions which occur in enclosure fires and lead to the freezing of CO burnout. Predictions of CO in the cooled layer are presented in the context of a complete computational fluid dynamics solution of velocity, temperature, and major species concentrations. A range of underhood equivalence ratios, from rich to lean, are investigated. The CMC method predicts CO in very good agreement with data. In particular, CMC is able to correctly predict CO concentrations in lean cooled gases, showing its capability in conditions where reaction rates change considerably.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

Carbon monoxide, the chief killer in fires, and other species are modelled for a series of enclosure fires. The conditions emulate building fires where CO is formed in the rich, turbulent, nonpremixed flame and is transported frozen to lean mixtures by the ceiling jet which is cooled by radiation and dilution. Conditional moment closure modelling is used and computational domain minimisation criteria are developed which reduce the computational cost of this method. The predictions give good agreement for CO and other species in the lean, quenched-gas stream, holding promise that this method may provide a practical means of modelling real, three-dimensional fire situations. (c) 2005 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

We consider the case of two cavity modes of the electromagnetic field, which are coupled via the action of a parametric amplifier. The fields are allowed to leak from the cavity and homodyne measurement is performed on one of the modes. Because of the correlations between the modes, this leads to a reduction of the variance in a quadrature of the other mode, although no measurement is performed on it directly. We discuss how this relates to the Einstein-Podolky-Rosen Gedankenexperiment.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

In a recent paper [16], one of us identified all of the quasi-stationary distributions for a non-explosive, evanescent birth-death process for which absorption is certain, and established conditions for the existence of the corresponding limiting conditional distributions. Our purpose is to extend these results in a number of directions. We shall consider separately two cases depending on whether or not the process is evanescent. In the former case we shall relax the condition that absorption is certain. Furthermore, we shall allow for the possibility that the minimal process might be explosive, so that the transition rates alone will not necessarily determine the birth-death process uniquely. Although we shall be concerned mainly with the minimal process, our most general results hold for any birth-death process whose transition probabilities satisfy both the backward and the forward Kolmogorov differential equations.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

We consider the relation between the conditional moment closure (CMC) and the unsteady flamelet model (FM). The CMC equations were originally constructed as global equations, while FM was derived asymptotically for a thin reaction zone. The recent tendency is to use FM-type equations as global equations. We investigate the possible consequences and suggest a new version of FM: coordinate-invariant FM (CIFM). Unlike FM, CIFM complies with conditional properties of the exact transport equations which are used effectively in CMC. We analyse the assumptions needed to obtain another global version of FM: representative interactive flamelets (RIF), from original FM and demonstrate that, in homogeneous turbulence, one of these assumptions is equivalent to the main CMC hypothesis.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

When examining a rock mass, joint sets and their orientations can play a significant role with regard to how the rock mass will behave. To identify joint sets present in the rock mass, the orientation of individual fracture planer can be measured on exposed rock faces and the resulting data can be examined for heterogeneity. In this article, the expectation-maximization algorithm is used to lit mixtures of Kent component distributions to the fracture data to aid in the identification of joint sets. An additional uniform component is also included in the model to accommodate the noise present in the data.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

We show that by making conditional measurements on the Einstein-Podolsky-Rosen (EPR) squeezed vacuum [T. Opatrny, G. Kurizki, and D.-G. Welsch, Phys. Rev. A 61, 032302 (2000)], one can improve the efficacy of teleportation for both the position-difference, momentum-sum, and number-difference, phase-sum continuous variable teleportation protocols. We investigate the relative abilities of the standard and conditional EPR states, and show that by conditioning we can improve the fidelity of teleportation of coherent states from below to above the (F) over bar =2/3 boundary, thereby achieving unambiguously quantum teleportation.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

Eight species of ectomycorrhizal (ECM) fungi in the genera Amanita. Gymnoboletus, Lactarius, and Russula were isolated from subtropical plant communities in eastem Australia. Two species were isolated from each of rainforest, Nothofagus forest, Eucalyptus forest, and Eucalyptus dominated wallum (heath) forest. These communities differ strongly in their soluble soil nitrogen (N) composition. The ability of the fungi to use inorganic (nitrate, ammonium) and organic (amide, peptide, protein) nitrogen sources was determined. As the fungi did not grow in liquid culture, a 'floating culture' technique was devised that allows hyphal growth on a screen floating on liquid medium. With some exceptions, fungal biomass production in floating culture closely reflected fungal growth on solid media assessed by total colony glucosamine content. Most isolates grown in floating culture had similar glucosamine concentrations on all N sources, with isolate specific concentrations ranging from 6 to 12 mug glucosamine g(-1) DW. However, Russula spp. had up to 1.7-fold higher glucosamine concentrations when growing with glutamine or ammonium compared to nitrate, glutathione or protein. Floating cultures supplied with 0.5, 1.5. 4.5, or 10 mm N mostly produced greatest biomass with 4.5 mM N. In vitro nitrate reductase activity (NRA) ranged from very low (0.03 mumol NO2- g(-1) fw h(-1)) in Russula sp. (wallum) to high (2.16 mumol NO2- g(-1) fw h(-1)) in Gymnoboletus sp. (rainforest) and mirrored the fungi's ability to use nitrate as a N source. All Russula spp. (wallum, Nothofagus and Eucalyptus forests), Lactarills sp, (rainforest) and.4manita sp. (wallum) utilized ammonium and glutamine but had little ability to use other N sources. In contrast,Amanita species (Nothofagus and Eucalyptus forests) grew on all N sources but produced most biomass with ammonium and glutamine. Only Gymnoboletus sp. (rainforest) showed similar growth with nitrate and ammonium as N sources. Fungal N source use was not associated with taxonomic groups, but is discussed in the context of soil N sources in the different habitats.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

We present a novel maximum-likelihood-based algorithm for estimating the distribution of alignment scores from the scores of unrelated sequences in a database search. Using a new method for measuring the accuracy of p-values, we show that our maximum-likelihood-based algorithm is more accurate than existing regression-based and lookup table methods. We explore a more sophisticated way of modeling and estimating the score distributions (using a two-component mixture model and expectation maximization), but conclude that this does not improve significantly over simply ignoring scores with small E-values during estimation. Finally, we measure the classification accuracy of p-values estimated in different ways and observe that inaccurate p-values can, somewhat paradoxically, lead to higher classification accuracy. We explain this paradox and argue that statistical accuracy, not classification accuracy, should be the primary criterion in comparisons of similarity search methods that return p-values that adjust for target sequence length.