Search

In forensic genetics, short tandem repeats (STRs) are used for human identification (HID). Degraded biological trace samples with low amounts of short DNA fragments (low-quality DNA samples) pose a challenge for STR typing. Predefined single nucleotide polymorphisms (SNPs) can be amplified on short PCR fragments and used to generate SNP profiles from low-quality DNA samples. However, the stochastic results from low-quality DNA samples may result in frequent locus drop-outs and insufficient numbers of SNP genotypes for convincing identification of individuals. Shotgun DNA sequencing potentially analyses all DNA fragments in a sample in contrast to the targeted PCR-based sequencing methods and may be applied to DNA samples of very low quality, like heavily compromised crime-scene samples and ancient DNA samples. Here, we developed a statistical model for shotgun sequencing, sequence alignment, and genotype calling. Results from replicated shotgun sequencing of buccal swab (high-quality samples) and hair samples (low-quality samples) were arranged in a genotype-call confusion matrix to estimate the calling error probability by maximum likelihood and Bayesian inference. We developed formulas for calculating the evidential weight as a likelihood ratio (LR) based on data from dynamically selected SNPs from shotgun DNA sequencing. The method accounts for potential genotyping errors. Different genotype quality filters may be applied to account for genotyping errors. An error probability of zero resulted in the forensically commonly used LR formula. When considering a single SNP marker's contribution to the LR, error probabilities larger than zero reduced the LR contribution of matching genotypes and increased the LR in the case of a mismatch. We developed an open-source R package, wgsLR, which implements the method, including estimating the calling error probability and calculating LR values., Comment: 25 pages, 9 figures

Amemiya (1973) proposed a ``consistent initial estimator'' for the parameters in a censored regression model with normal errors. This paper demonstrates that a similar approach can be used to construct moment conditions for fixed--effects versions of the model considered by Amemiya. This result suggests estimators for models that have not previously been considered., Comment: This is a very old unpublished draft that is relevant for the recent literature that constructs moment conditions in nonlinear panel data models

We consider reconstruction of an ambient signal in a compressed sensing (CS) setup where the ambient signal has a neural network based generative model. The generative model has a sparse-latent input and we refer to the generated ambient signal as generative sparse-latent signal (GSL). The proposed sparsity inducing reconstruction algorithm is inherently non-convex, and we show that a gradient based search provides a good reconstruction performance. We evaluate our proposed algorithm using simulated data., Comment: Accepted at 31st European Signal Processing Conference, EUSIPCO 2023

In-order to understand the results of recent observations of exoplanets, models have become increasingly complex. Unfortunately this increases both the computational cost and output size of said models. We intend to explore if AI-image-recognition can alleviate this burden. We used DYNAMICO to run a series of HD209458-like models with different orbital-radii. Training data for a number of features of interest was selected from the initial outputs of these models. This was used to train a pair of multi-categorisation convolutional-neural-networks (CNN), which we applied to our outer-atmosphere-equilibrated models. The features detected by our CNNs revealed that our models fall into two regimes: models with a shorter orbital-radii exhibit significant global mixing which shapes the entire atmospheres dynamics. Whereas, models with longer orbital-radii exhibit negligible mixing except at mid-pressures. Here, the initial non-detection of any trained features revealed a surprise: a night-side hot-spot. Analysis suggests that this occurs when rotational influence is sufficiently weak that divergent flows from the day-side to the night-side dominate over rotational-driven transport, such as the equatorial jet. We suggest that image-classification may play an important role in future, computational, atmospheric studies. However special care must be paid to the data feed into the model, from the colourmap, to training the CNN on features with enough breadth and complexity that the CNN can learn to detect them. However, by using preliminary-studies and prior-models, this should be more than achievable for future exascale calculations, allowing for a significant reduction in future workloads and computational resources., Comment: Accepted for publication in ApJ

We address the tasks of Bayesian state estimation and forecasting for a model-free process in an unsupervised learning setup. For a model-free process, we do not have any a-priori knowledge of the process dynamics. In the article, we propose DANSE -- a Data-driven Nonlinear State Estimation method. DANSE provides a closed-form posterior of the state of the model-free process, given linear measurements of the state. In addition, it provides a closed-form posterior for forecasting. A data-driven recurrent neural network (RNN) is used in DANSE to provide the parameters of a prior of the state. The prior depends on the past measurements as input, and then we find the closed-form posterior of the state using the current measurement as input. The data-driven RNN captures the underlying non-linear dynamics of the model-free process. The training of DANSE, mainly learning the parameters of the RNN, is executed using an unsupervised learning approach. In unsupervised learning, we have access to a training dataset comprising only a set of measurement data trajectories, but we do not have any access to the state trajectories. Therefore, DANSE does not have access to state information in the training data and can not use supervised learning. Using simulated linear and non-linear process models (Lorenz attractor and Chen attractor), we evaluate the unsupervised learning-based DANSE. We show that the proposed DANSE, without knowledge of the process model and without supervised learning, provides a competitive performance against model-driven methods, such as the Kalman filter (KF), extended KF (EKF), unscented KF (UKF), a data-driven deep Markov model (DMM) and a recently proposed hybrid method called KalmanNet. In addition, we show that DANSE works for high-dimensional state estimation., Comment: 12 pages, Accepted for publication in IEEE Transactions in Signal Processing

In this paper we consider a geothermal energy storage in which the spatio-temporal temperature distribution is modeled by a heat equation with a convection term. Such storages often are embedded in residential heating systems and control and management require the knowledge of some aggregated characteristics of that temperature distribution in the storage. They describe the input-output behaviour of the storage and the associated energy flows and their response to charging and discharging processes. We aim to derive an efficient approximative description of these characteristics by a low-dimensional system of ODEs. This leads to a model order reduction problem for a large scale linear system of ODEs arising from the semi-discretization of the heat equation combined with a linear algebraic output equation. In a first step we approximate the non time-invariant system of ODEs by a linear time-invariant system. Then we apply Lyapunov balanced truncation model order reduction to approximate the output by a reduced-order system with only a few state equations but almost the same input-output behavior. The paper presents results of extensive numerical experiments showing the efficiency of the applied model order reduction methods. It turns out that only a few suitable chosen ODEs are sufficient to produce good approximations of the input-output behaviour of the storage., Comment: arXiv admin note: text overlap with arXiv:2104.05116

For the parabolic and the transport equations whose coefficient of the first term, denoted by $b$, can be in a negative Besov space,we provide a control of the solution in the H\"older space $C^\gamma$, $\gamma \in (0,1)$, regardless of $b$ and matchingwith the source functions regularity. We define some kind of solutions which do not require $b$ to be Lipschitz continuous. For a ``very weak'' solution, called in the paper \textit{mild vanishing viscous}, in $C^\gamma$, there is no regularity constraint on $b$. If $b$ lies in a $\tilde \gamma$-H\"older space, $\tilde \gamma>1-\gamma$, we establish that there is a weak solution in a $\gamma$-H\"older space. If $b$ is supposed to be divergence free, we obtain the same result for $b$ having a negative regularity in space, precisely in $L^\infty(B_{\infty,\infty}^{-\beta})$ for $\beta<\gamma$. Finally, if $\tilde \gamma > \frac{1}{1+\gamma}$with a \textit{vanishing viscous} condition, then the selected solution is unique. In this case, there is somehow a \textit{regularisation} by turbulence (corresponding to the Reynolds number going to $+ \infty$); the \textit{vanishing viscosity} overwhelms the potential blowing up of the rough coefficients. Importantly, as a by-product of our analysis, we are able to give a meaning of a product of distributions. For $b$ lying in a $C^\gamma$, we obtain the same condition as for the usual Bony's para-product; but in a weaker solution framework, the product is defined beyond the para-product condition and even with no constraint at all in the \textit{mild vanishing viscous} context. We also obtain that the time averaging of the distributions product is $\gamma$-H\"older continuous. These new results happens because, in the considered product, one of the distribution is the gradient of a solution of a Partial Differential Equation. Thanks to our analysis, we also get a H\"older control of a solution of the inviscid Burgers' equation. Under some regularity and vanishing viscous constraint, the solution is a weak solution and is unique in a certain sense. The vanishing viscous procedure allows to avoid the well-known critical time of the solution built by characteristics.

We obtain the well-posedness and Schauder estimates for a class of system of linear, quasi-linear and non-linear second order partial differential equations. We deduce existence and uniqueness of a global smooth solution of a non-linear and non-local equation that we call "semi" incompressible Navier Stokes equation in R 3 .

Two of Peter Schmidt's many contributions to econometrics have been to introduce a simultaneous logit model for bivariate binary outcomes and to study estimation of dynamic linear fixed effects panel data models using short panels. In this paper, we study a dynamic panel data version of the bivariate model introduced in Schmidt and Strauss (1975) that allows for lagged dependent variables and fixed effects as in Ahn and Schmidt (1995). We combine a conditional likelihood approach with a method of moments approach to obtain an estimation strategy for the resulting model. We apply this estimation strategy to a simple model for the intra-household relationship in employment. Our main conclusion is that the within-household dependence in employment differs significantly by the ethnicity composition of the couple even after one allows for unobserved household specific heterogeneity.

The amount of data generated by numerical simulations in various scientific domains such as molecular dynamics, climate modeling, biology, or astrophysics, led to a fundamental redesign of application workflows. The throughput and the capacity of storage subsystems have not evolved as fast as the computing power in extreme-scale supercomputers. As a result, the classical post-hoc analysis of simulation outputs became highly inefficient. In-situ workflows have then emerged as a solution in which simulation and data analytics are intertwined through shared computing resources, thus lower latencies. Determining the best allocation, i.e., how many resources to allocate to each component of an in-situ workflow; and mapping, i.e., where and at which frequency to run the data analytics component, is a complex task whose performance assessment is crucial to the efficient execution of in-situ workflows. However, such a performance evaluation of different allocation and mapping strategies usually relies either on directly running them on the targeted execution environments, which can rapidly become extremely time-and resource-consuming, or on resorting to the simulation of simplified models of the components of an in-situ workflow, which can lack of realism. In both cases, the validity of the performance evaluation is limited. To address this issue, we introduce SIM-SITU, a framework for the faithful simulation of in-situ workflows. This framework builds on the SimGrid toolkit and benefits of several important features of this versatile simulation tool. We designed SIM-SITU to reflect the typical structure of in-situ workflows and thanks to its modular design, SIM-SITU has the necessary flexibility to easily and faithfully evaluate the behavior and performance of various allocation and mapping strategies for in-situ workflows. We illustrate the simulation capabilities of SIM-SITU on a Molecular Dynamics use case. We study the impact of different allocation and mapping strategies on performance and show how users can leverage SIM-SITU to determine interesting tradeoffs when designing their in-situ workflow.

Magnetic tapes are often considered as an outdated storage technology, yet they are still used to store huge amounts of data. Their main interests are a large capacity and a low price per gigabyte, which come at the cost of a much larger file access time than on disks. With tapes, finding the right ordering of multiple file accesses is thus key to performance. Moving the reading head back and forth along a kilometer long tape has a non-negligible cost and unnecessary movements thus have to be avoided. However, the optimization of tape request ordering has then rarely been studied in the scheduling literature, much less than I/O scheduling on disks. For instance, minimizing the average service time for several read requests on a linear tape remains an open question. Therefore, in this paper, we aim at improving the quality of service experienced by users of tape storage systems, and not only the peak performance of such systems. To this end, we propose a reasonable polynomial-time exact algorithm while this problem and simpler variants have been conjectured NP-hard. We also refine the proposed model by considering U-turn penalty costs accounting for inherent mechanical accelerations. Then, we propose low-cost variants of our optimal algorithm by restricting the solution space, yet still yielding an accurate suboptimal solution. Finally, we compare our algorithms to existing solutions from the literature on logs of the mass storage management system of a major datacenter. This allows us to assess the quality of previous solutions and the improvement achieved by our low-cost algorithms. Aiming for reproducibility, we make available the complete implementation of the algorithms used in our evaluation, alongside the dataset of tape requests that is, to the best of our knowledge, the first of its kind to be publicly released.

Data augmentation is an important component in the robustness evaluation of models in natural language processing (NLP) and in enhancing the diversity of the data they are trained on. In this paper, we present NL-Augmenter, a new participatory Python-based natural language augmentation framework which supports the creation of both transformations (modifications to the data) and filters (data splits according to specific features). We describe the framework and an initial set of 117 transformations and 23 filters for a variety of natural language tasks. We demonstrate the efficacy of NL-Augmenter by using several of its transformations to analyze the robustness of popular natural language models. The infrastructure, datacards and robustness analysis results are available publicly on the NL-Augmenter repository (https://github.com/GEM-benchmark/NL-Augmenter)., Comment: 39 pages, repository at https://github.com/GEM-benchmark/NL-Augmenter

One factor of success in software development companies is their ability to deliver good quality products, fast. For this, they need to improve their software development practices. We work with a medium-sized company modernizing its development practices. The company introduced several practices recommended in agile development. If the benefits of these practices are well documented, the impact of such changes on the developers is less well known. We follow this modernization before and during the COVID-19 outbreak. This paper presents an empirical study of the perceived benefit and drawback of these practices as well as the impact of COVID-19 on the company's employees. One of the conclusions, is the additional difficulties created by obsolete technologies to adapt the technology itself and the development practices it encourages to modern standards., Comment: 10 pages, 3 figures, 2 tables

This paper studies a dynamic ordered logit model for panel data with fixed effects. The main contribution of the paper is to construct a set of valid moment conditions that are free of the fixed effects. The moment functions can be computed using four or more periods of data, and the paper presents sufficient conditions for the moment conditions to identify the common parameters of the model, namely the regression coefficients, the autoregressive parameters, and the threshold parameters. The availability of moment conditions suggests that these common parameters can be estimated using the generalized method of moments, and the paper documents the performance of this estimator using Monte Carlo simulations and an empirical illustration to self-reported health status using the British Household Panel Survey.

In pursuit of explainability, we develop generative models for sequential data. The proposed models provide state-of-the-art classification results and robust performance for speech phone classification. We combine modern neural networks (normalizing flows) and traditional generative models (hidden Markov models - HMMs). Normalizing flow-based mixture models (NMMs) are used to model the conditional probability distribution given the hidden state in the HMMs. Model parameters are learned through judicious combinations of time-tested Bayesian learning methods and contemporary neural network learning methods. We mainly combine expectation-maximization (EM) and mini-batch gradient descent. The proposed generative models can compute likelihood of a data and hence directly suitable for maximum-likelihood (ML) classification approach. Due to structural flexibility of HMMs, we can use different normalizing flow models. This leads to different types of HMMs providing diversity in data modeling capacity. The diversity provides an opportunity for easy decision fusion from different models. For a standard speech phone classification setup involving 39 phones (classes) and the TIMIT dataset, we show that the use of standard features called mel-frequency-cepstral-coeffcients (MFCCs), the proposed generative models, and the decision fusion together can achieve $86.6\%$ accuracy by generative training only. This result is close to state-of-the-art results, for examples, $86.2\%$ accuracy of PyTorch-Kaldi toolkit [1], and $85.1\%$ accuracy using light gated recurrent units [2]. We do not use any discriminative learning approach and related sophisticated features in this article., Comment: 12 pages, 4 figures

This paper is devoted to numerical simulations of the short-term behavior of the spatial temperature distribution in a geothermal energy storage. Such simulations are needed for the optimal control and management of residential heating systems equipped with an underground thermal storage. We apply numerical methods derived in our companion paper [15] in which we study the governing initial boundary value problem for a linear heat equation with convection. Further, we perform extensive numerical experiments in order to investigate properties of the spatio-temporal temperature distribution and of its aggregated characteristics., Comment: 29 pages. arXiv admin note: text overlap with arXiv:2104.05005

This paper investigates numerical methods for simulations of the short-term behavior of a geothermal energy storage. Such simulations are needed for the optimal control and management of residential heating systems equipped with an underground thermal storage. There a given volume under or aside of a building is filled with soil and insulated to the surrounding ground. The thermal energy is stored by raising the temperature of the soil inside the storage. It is charged and discharged via pipe heat exchangers filled with a moving fluid. Simulations of geothermal energy storages aim to determine how much energy can be stored in or taken from the storage within a given short period of time. The latter depends on the dynamics of the spatial temperature distribution in the storage which is governed by a linear heat equation with convection and appropriate boundary and interface conditions. We consider semi- and full discretization of that PDE using finite difference schemes and study associated stability problems. Numerical results based on the derived methods are presented in the companion paper [17]., Comment: 26 pages

We test the robustness of a maximum-likelihood (ML) based classifier where sequential data as observation is corrupted by noise. The hypothesis is that a generative model, that combines the state transitions of a hidden Markov model (HMM) and the neural network based probability distributions for the hidden states of the HMM, can provide a robust classification performance. The combined model is called normalizing-flow mixture model based HMM (NMM-HMM). It can be trained using a combination of expectation-maximization (EM) and backpropagation. We verify the improved robustness of NMM-HMM classifiers in an application to speech recognition., Comment: 6 pages. Accepted at MLSP 2020

This paper investigates the construction of moment conditions in discrete choice panel data with individual specific fixed effects. We describe how to systematically explore the existence of moment conditions that do not depend on the fixed effects, and we demonstrate how to construct them when they exist. Our approach is closely related to the numerical "functional differencing" construction in Bonhomme (2012), but our emphasis is to find explicit analytic expressions for the moment functions. We first explain the construction and give examples of such moment conditions in various models. Then, we focus on the dynamic binary choice logit model and explore the implications of the moment conditions for identification and estimation of the model parameters that are common to all individuals.

We explore the use of traditional and contemporary hidden Markov models (HMMs) for sequential physiological data analysis and sepsis prediction in preterm infants. We investigate the use of classical Gaussian mixture model based HMM, and a recently proposed neural network based HMM. To improve the neural network based HMM, we propose a discriminative training approach. Experimental results show the potential of HMMs over logistic regression, support vector machine and extreme learning machine., Comment: Submitted at the 45th International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2020, Barcelona, Spain

Hidden Markov model (HMM) has been successfully used for sequential data modeling problems. In this work, we propose to power the modeling capacity of HMM by bringing in neural network based generative models. The proposed model is termed as GenHMM. In the proposed GenHMM, each HMM hidden state is associated with a neural network based generative model that has tractability of exact likelihood and provides efficient likelihood computation. A generative model in GenHMM consists of mixture of generators that are realized by flow models. A learning algorithm for GenHMM is proposed in expectation-maximization framework. The convergence of the learning GenHMM is analyzed. We demonstrate the efficiency of GenHMM by classification tasks on practical sequential data. Code available at https://github.com/FirstHandScientist/genhmm.

This paper introduces measures for how each moment contributes to the precision of parameter estimates in GMM settings. For example, one of the measures asks what would happen to the variance of the parameter estimates if a particular moment was dropped from the estimation. The measures are all easy to compute. We illustrate the usefulness of the measures through two simple examples as well as an application to a model of joint retirement planning of couples. We estimate the model using the UK-BHPS, and we find evidence of complementarities in leisure. Our sensitivity measures illustrate that the estimate of the complementarity is primarily informed by the distribution of differences in planned retirement dates. The estimated econometric model can be interpreted as a bivariate ordered choice model that allows for simultaneity. This makes the model potentially useful in other applications.