Efficient Data Augmentation for Fitting Stochastic Epidemic Models to Prevalence Data

Stochastic epidemic models describe the dynamics of an epidemic as a disease spreads through a population. Typically, only a fraction of cases are observed at a set of discrete times. The absence of complete information about the time evolution of an epidemic gives rise to a complicated latent variable problem in which the state space size of the epidemic grows large as the population size increases. This makes analytically integrating over the missing data infeasible for populations of even moderate size. We present a data augmentation Markov chain Monte Carlo (MCMC) framework for Bayesian estimation of stochastic epidemic model parameters, in which measurements are augmented with subject-level disease histories. In our MCMC algorithm, we propose each new subject-level path, conditional on the data, using a time-inhomogenous continuous-time Markov process with rates determined by the infection histories of other individuals. The method is general, and may be applied to a broad class of epidemic models with only minimal modifications to the model dynamics and/or emission distribution. We present our algorithm in the context of multiple stochastic epidemic models in which the data are binomially sampled prevalence counts, and apply our method to data from an outbreak of influenza in a British boarding school. Supplementary material for this article is available online.     

Heat transfer augmentation and pumping power in double-pipe heat exchangers

One of the criteria for evaluating the performance of a heat exchanger with extended surfaces is the pumping power required for a specified heat duty. The results of an experimental project to relate the pumping power to heat transfer augmentation in a double-pipe heat exchanger are reported. The inner, electrically heated pipe was provided with external, rectangular, axial extended surfaces with interruptions. Heat transfer augmentation and friction factors were determined for different configurations with air as the fluid. Starting with continuous fins, cuts were introduced in the fins to give four ratios of the finssegment length to the gap between the segments, and finally all the fins were removed, which resulted in smooth pipes. Five different mass flow rates in two different inner pipes were employed. Lengths, surface areas, and pumping powers for finned pipes are compared with those for smooth pipes. The average heat transfer coefficient increases with an increase in the frequency of the interruptions. For equal heat transfer rates a significant reduction in the lengths can be achieved by interrupted fins. In many cases the reduction in the length is also accompanied by a reduction in the pumping power.     

Experimental study of heat and mass transfer from a horizontal cylinder in downward air-water mist flow with blockage effect

An experimental study was made on convective heat and mass transfer from a horizontal heated cylinder in a downward flow of air-water mist at a blockage ratio of 0.4. The measured local heat transfer coefficients agree fairly well with the authors' numerical solutions obtained previously for the front surface of a cylinder over the ranges mass flow ratio 0–4.5×10⁻², a temperature difference between the cylinder and air 10–43 K, gas Reynolds number (7.9–23)×10³, Rosin-Rammler size parameter 105–168 μm, and dispersion parameter 3.4–3.7. Heat transfer augmentation, two-pahse to single-phase of greater than 19 was attained at the forward stagnation point. For heat transfer in the rear part of the cylinder, an empirical formula is derived by taking into account the dimensionless governing variables, that is, coolant-feed and evaporation parameters.     

A Bayesian approach for analyzing case 2 interval-censored data under the semiparametric proportional odds model

Analyzing interval-censored data is difficult due to its complex data structure containing left-, interval-, and right-censored observations. An easy-to-implement Bayesian approach is proposed under the proportional odds (PO) model for analyzing such data. The nondecreasing baseline log odds function is modeled with a linear combination of monotone splines. Two efficient Gibbs samplers are developed based on two different data augmentations using the relationship between the PO model and the logistic distribution. In the first data augmentation, the logistic distribution is achieved by the scaled normal mixture with the scale parameter related to the Kolmogorov-Smirnove distribution. In the second data augmentation, the logistic distribution is approximated by a Student’s t distribution up to a scale constant. The proposed methods are evaluated by simulation studies and illustrated with an application of an HIV data set.     

基于DCGAN的拉曼光谱样本扩充及应用研究

拉曼光谱检测方法依赖于化学计量学算法,深度学习是当下最炙手可热的方向,可应用于拉曼光谱进行建模。但是深度学习需要大样本进行训练,而拉曼光谱采集受制于器材和人力成本,获取大批量的样本需要更大成本,且易受荧光等因素干扰,这些问题都制约了将深度学习应用于拉曼光谱。针对以上问题,通过引入深度卷积生成对抗网络（DCGAN）提取拉曼光谱内部特征,对抗生成新的拉曼光谱,从而达到扩充数据集目的。同时和另一个扩充数据集的方法--偏移法进行对比,证明DCGAN的可靠性。设计生成光谱选取标准,选取高相似性的光谱填充数据集,为深度学习在拉曼光谱中的应用奠定基础。为了验证生成的光谱比原始光谱有更好的适用性,设计四组实验：（1）使用原始拉曼光谱输入到SVM进行分类,得到51.92%的分类准确率;（2）使用原始拉曼光谱输入到CNN进行分类,得到75.00%的分类准确率;（3）采用偏移法生成光谱,输入到CNN里进行分类,得到91.85%的分类准确率;（4）使用DCGAN生成光谱,输入到CNN里进行分类,得到98.52%分类准确率。实验结果表明,DCGAN能在只有少量拉曼光谱的情况下,通过对抗学习得到较好的生成光谱,且生成的光谱相比原光谱更加清晰,减少了可能的干扰因素,具有光谱预处理效果。通过DCGAN对抗生成大量高质量的数据填充到原有拉曼光谱数据集,扩充数据集的样本量,使得深度学习模型能够得到更好的训练,从而提高模型的准确率。该研究为深度学习方法应用于拉曼光谱分析技术提出了一个可行的方案。     

Adding more observations to saturated D‐optimal resolution III two‐level factorial designs

We consider the class of saturated main effect plans for the 2^k factorial. With these saturated designs, the overall mean and all main effects can be unbiasedly estimated provided that there are no interactions. However, there is no way to estimate the error variance with such designs. Because of this and other reasons, we like to add some additional runs to the set of (k+1) runs in the D‐optimal design in this class. Our goals here are: (1) to search for s additional runs so that the resulting design based on (k+s+1) runs yields a D‐optimal design in the class of augmented designs; (2) to classify all the runs into equivalent classes so that the runs in the same equivalent class give us the same value of the determinant of the information matrix. This allows us to trade runs for runs if this becomes necessary; (3) to obtain upper bounds for determinant of the information matrices of augmented designs. In this article we shall address these approaches and present some new results. © 2002 Wiley Periodicals, Inc. J Combin Designs 11: 51–77, 2003; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/jcd.10026     

The Stable Class of the Augmentation Ideal

In [F.E.A. Johnson, Stable Modules and the D(2)-Problem, LMS Lecture Notes In Mathematics, vol. 301, CUP (2003)], for finite groups G, we gave a parametrization of the stable class of the augmentation ideal of Z[G] in terms of stably free modules. Whilst the details of this parametrization break down immediately for infinite groups, nevertheless one may hope to find parallel arguments for restricted classes of infinite groups. Subject to the restriction that Ext¹(Z, Z[G]) = 0, we parametrize the minimal level in Ω₁(Z) by means of stably free modules and give a lower estimate for the size of Ω₁(Z).     

The Self-Multiset Sampler

The multiset sampler has been shown to be an effective algorithm to sample from complex multimodal distributions, but the multiset sampler requires that the parameters in the target distribution can be divided into two parts: the parameters of interest and the nuisance parameters. We propose a new self-multiset sampler (SMSS), which extends the multiset sampler to distributions without nuisance parameters. We also generalize our method to distributions with unbounded or infinite support. Numerical results show that the SMSS and its generalization have a substantial advantage in sampling multimodal distributions compared to the ordinary Markov chain Monte Carlo algorithm and some popular variants. Supplemental materials for the article are available online.