Likelihood Distribution Research Articles

Intelligent geospatial maritime risk analytics using the Discrete Global Grid System

ABSTRACT Each year, accidents involving ships result in significant loss of life, environmental pollution and economic losses. The promotion of navigation safety through risk reduction requires methods to assess the spatial distribution of the relative likelihood of occurrence. Yet, such methods necessitate the integration of large volumes of heterogenous datasets which are not well suited to traditional data structures. This paper proposes the use of the Discrete Global Grid System (DGGS) as an efficient and advantageous structure to integrate vessel traffic, metocean, bathymetric, infrastructure and other relevant maritime datasets to predict the occurrence of ship groundings. Massive and heterogenous datasets are well suited for machine learning algorithms and this paper develops a spatial maritime risk model based on a DGGS utilising such an approach. A Random Forest algorithm is developed to predict the frequency and spatial distribution of groundings while achieving an R2 of 0.55 and a mean squared error of 0.002. The resulting risk maps are useful for decision-makers in planning the allocation of mitigation measures, targeted to regions with the highest risk. Further work is identified to expand the applications and insights which could be achieved through establishing a DGGS as a global maritime spatial data structure.

Fourier-Analysis-Based Form of Normalized Maximum Likelihood: Exact Formula and Relation to Complex Bayesian Prior

Normalized maximum likelihood (NML) distribution of probabilistic model gives the optimal code length function in the sense of minimax regret. Despite its optimal property, the calculation of NML distribution is not easy, and existing efficient methods have been focusing on its asymptotic behavior, or on specific models. This paper gives an efficient way to calculate NML by integral on parameter domain, not on data domain, showing that NML distribution is a Bayesian predictive distribution with a complex prior, based on our novel Fourier expansion approach. Our results provide an integrated way to calculate NML for exponential family and also include a non-asymptotic version of previous work on asymptotic behavior for general cases. The applications of our methodology are not limited to but also include normal distribution, Gamma distribution, Weibull distribution, and von Mises distribution.

Assessment of Sound Source Tracking Using Multiple Drones Equipped with Multiple Microphone Arrays.

Drone audition techniques are helpful for listening to target sound sources from the sky, which can be used for human searching tasks in disaster sites. Among many techniques required for drone audition, sound source tracking is an essential technique, and thus several tracking methods have been proposed. Authors have also proposed a sound source tracking method that utilizes multiple microphone arrays to obtain the likelihood distribution of the sound source locations. These methods have been demonstrated in benchmark experiments. However, the performance against various sound sources with different distances and signal-to-noise ratios (SNRs) has been less evaluated. Since drone audition often needs to listen to distant sound sources and the input acoustic signal generally has a low SNR due to drone noise, making a performance assessment against source distance and SNR is essential. Therefore, this paper presents a concrete evaluation of sound source tracking methods using numerical simulation, focusing on various source distances and SNRs. The simulated results captured how the tracking performance will change when the sound source distance and SNR change. The proposed approach based on location distribution estimation tended to be more robust against distance increase, while existing approaches based on directional estimation tended to be more robust against decreasing SNR.

Seismic Phase Association Based on the Maximum Likelihood Method

Phase association is a process that links seismic phases triggered at the stations of a seismic network to declare the occurrence of earthquakes. During phase association, a set of phases from different stations is examined to determine the common origin of phases within a specific region, predominantly on the basis of a grid search and the sum of observations. The association of seismic phases in local earthquake monitoring systems or earthquake early warning systems is often disturbed not only by transient noises, but also by large regional or teleseismic events. To mitigate this disturbance, we developed a seismic phase association method, binder_max, which uses the maximum likelihood method to associate seismic phases. The method is based on the framework of binder_ew, the phase associator of Earthworm, but it uses a likelihood distribution of the arrival information instead of stacking arrival information. Applying binder_max to data from seismic networks of South Korea and Ohio, United States, we found a significant improvement in the robustness of the method against large regional or teleseismic events compared to binder_ew. Our results indicate that binder_max can associate seismic phases of local earthquakes to the same degree as binder_ew as well as can avoid many of the false associations that have limited binder_ew.

Land-Use Regression of Long-Term Transportation Data on Metabolic Syndrome Risk Factors in Low-Income Communities

Traffic-related air pollution has been associated with adverse cardiovascular health effects in near-road residents. Transportation parameters are important surrogate variables to determine spatial variation of air pollution and consequential health outcomes. We used land-use regression models to explore associations between cardiovascular (metabolic syndrome [MetS]) health outcomes collected from a sample of low-income participants ( N = 4,959) and transportation parameters within a defined impact zone of a participant’s residence. We hypothesize cardiovascular risk factors are associated with spatially distributed transportation parameters and land-use data. MetS risk factors (waist circumference, blood pressure, triglycerides, HDL-cholesterol, and glucose) were obtained from 4,945 participants between 2014 and 2020 across the city of El Paso, Texas. Traffic-related and land-use variables were acquired from the El Paso MPO and the U.S. Census Bureau within two impact zones of 500 m and 1,000 m radius, centered at each participant resident’s home latitude and longitude coordinates using GIS. The increase in street length within 500 m radius was found to associate with increases in BMI, waist circumference, triglycerides, and glucose ( p < 0.05). Glucose showed positive relationships with inverse distance to the nearest international ports of entry ( p = 0.02). Also, as the total length of the street increases, the likelihood of a high waist ( p < 0.01), high triglycerides ( p = 0.03), and low HDL-cholesterol ( p < 0.01) also increased. Based on a multivariable regression model, a probability surface map was prepared to show the spatial distribution of likelihood for acquiring MetS in El Paso, TX.

SateLoc: A Virtual Fingerprinting Approach to Outdoor LoRa Localization Using Satellite Images

With the increasing relevance of the Internet of Things and large-scale location-based services, LoRa localization has been attractive due to its low-cost, low-power, and long-range properties. However, existing localization approaches based on received signal strength indicators are either easily affected by signal fading of different land-cover types or labor intensive. In this work, we propose SateLoc, a LoRa localization system that utilizes satellite images to generate virtual fingerprints. Specifically, SateLoc first uses high-resolution satellite images to identify land-cover types. With the path loss parameters of each land-cover type, SateLoc can automatically generate a virtual fingerprinting map for each gateway. We then propose a novel multi-gateway combination strategy, which is weighted by the environmental interference of each gateway, to produce a joint likelihood distribution for localization and tracking. We implement SateLoc with commercial LoRa devices without any hardware modification, and evaluate its performance in a 227,500-m urban area. Experimental results show that SateLoc achieves a median localization error of 43.5 m, improving more than 50% compared to state-of-the-art model-based approaches. Moreover, SateLoc can achieve a median tracking error of 37.9 m with the distance constraint of adjacent estimated locations. More importantly, compared to fingerprinting-based approaches, SateLoc does not require the labor-intensive fingerprint acquisition process.

Calibration of Spalart-Allmaras model for simulation of corner flow separation in linear compressor cascade

This study focuses on the calibration of Spalart--Allmaras turbulence model parameters using the Bayesian inference approach to reproduce experimental measurements of corner flow separation in linear compressor cascade. The quantity of interest selected for the calibration process is the pitchwise distribution of Mach number in the wake of the linear compressor cascade. The model parameters are assumed to be random variables obeying uniform prior probability distributions. Sensitivity analysis is used to rank the importance and select the most influential turbulence model parameters for the calibration process. The sensitivity ranking indicates that two model parameters cb1 and kappa are the most influential random variables resulting in a two--parameter Bayesian calibration process. The likelihood distribution is specified in the form of the Gauss distribution to include the experimental uncertainty. The likelihood distribution is used together with prior distribution to compute posterior probabilities of selected model parameters. The polynomial chaos expansion is employed as a surrogate model to reduce the cost of posterior calculation. Numerical simulations with calibrated turbulence parameters show a significant increase in the accuracy of Mach number profile prediction for separated flows in linear compressor cascade. Numerical simulations also demonstrate that the calibrated set of model coefficients produce accurate predictions of the total pressure and Mach number profiles for the range of incidence angles that were not part of the calibration process.

A new parametric generalized exponential entropy measure on intuitionistic vague sets

Entropy is a significant mathematical instrument for determining ambiguous/fuzzy information. Entropy is indispensable for measuring ambiguity, first familiarized by Shannon (Syst Tech J 27:379–423, 1948) to extent the degree of uncertainty in likelihood distributions. Complex info processes are extensively pragmatic in decision-making processes. Created as per the notion of an exponential exponent for the fuzzy set, our exertion offers a measure of the power of an intuitive set of exponents. This article identifies a new measure of the exponential theorem on an intuitive set of equations. In addition, the necessary properties are displayed. By analyzing the results of the examples, it has been shown that this method is faster and more effective in practice.

Probabilistic dipole inversion for adaptive quantitative susceptibility mapping

A learning-based posterior distribution estimation method, Probabilistic Dipole Inversion (PDI), is proposed to solve the quantitative susceptibility mapping (QSM) inverse problem in MRI with uncertainty estimation. In PDI, a deep convolutional neural network (CNN) is used to represent the multivariate Gaussian distribution as the approximate posterior distribution of susceptibility given the input measured field. Such CNN is first trained on healthy subjects via posterior density estimation, where the training dataset contains samples from the true posterior distribution. Domain adaptations are then deployed on patient datasets with new pathologies not included in pre-training, where PDI updates the pre-trained CNN’s weights in an unsupervised fashion by minimizing the Kullback-Leibler divergence between the approximate posterior distribution represented by CNN and the true posterior distribution from the likelihood distribution of a known physical model and pre-defined prior distribution. Based on our experiments, PDI provides additional uncertainty estimation compared to the conventional MAP approach, while addressing the potential issue of the pre-trained CNN when test data deviates from training. Our code is available at https://github.com/Jinwei1209/Bayesian_QSM.

Body site distribution of pediatric-onset morphea and association with extracutaneous manifestations

The distribution of pediatric-onset morphea and site-based likelihood for extracutaneous complications has not been well characterized. To characterize the lesional distribution of pediatric-onset morphea and to determine the sites with the highest association of extracutaneous manifestations. A retrospective cross-sectional study was performed. Using clinical photographs, morphea lesions were mapped onto body diagrams using customized software. A total of 823 patients with 2522 lesions were included. Lesions were more frequent on the superior (vs inferior) anterior aspect of the head and extensor (vs flexor) extremities. Linear morphea lesions were more likely on the head and neck, whereas plaque and generalized morphea lesions were more likely on the trunk. Musculoskeletal complications were more likely with lesions on the extensor (vs flexor) extremity (odds ratio [OR], 2.0; 95% confidence interval [CI], 1.2-3.4), whereas neurologic manifestations were more likely with lesions on the anterior (vs posterior) (OR, 2.8; 95% CI, 1.7-4.6) and superior (vs inferior) aspect of the head (OR, 2.3; 95% CI, 1.6-3.4). Retrospective nature and the inclusion of only patients with clinical photographs. The distribution of pediatric-onset morphea is not random and varies with body site and within individual body sites. The risk stratification of extracutaneous manifestations by body site may inform decisions about screening for extracutaneous manifestations, although prospective studies are needed.

Model selection and parameter estimation using the iterative smoothing method

We compute the distribution of likelihoods from the non-parametric iterative smoothing method over a set of mock Pantheon-like type Ia supernova datasets. We use this likelihood distribution to test whether typical dark energy models are consistent with the data and to perform parameter estimation. In this approach, the consistency of a model and the data is determined without the need for comparison with another alternative model. Simulating future WFIRST-like data, we study type II errors and show how confidently we can distinguish different dark energy models using this non-parametric approach.

Genetic Population Structure of Wild Pigs in Southern Texas.

Simple SummaryWild pigs are the most abundant wild exotic ungulate in the United States. In Texas, particularly, they are abundant and represent a threat to ecosystems, agriculture and humans. Our objective was to apply a landscape-scale analysis of population genetic structure of wild pigs to aid in their management in southern Texas. We used microsatellites to assist large-scale applied management. We found that some populations were isolated from one another. However, many individuals and local populations were admixed, which indicates that multiple introductions and artificial movement of individuals has occurred. Wild pig management efficiency and effectiveness may be able to improve if illegal translocations stop (e.g., enforcing laws) and if management cooperatives are created to manage spatially extensive areas of southern Texas.Wild pigs (Sus scrofa) alter ecosystems, affect the economy, and carry diseases that can be transmitted to livestock, humans, and wildlife. Understanding wild pig movements and population structure data, including natural population boundaries and dispersal, may potentially increase the efficiency and effectiveness of management actions. We trapped, conducted aerial shootings, and hunted wild pigs from 2005 to 2009 in southern Texas. We used microsatellites to assist large-scale applied management. We quantify broad-scale population structure among 24 sites across southern Texas by computing an overall FST value, and a Bayesian clustering algorithm both with and without considering the spatial location of samples. At a broad geographic scale, pig populations displayed a moderate degree of genetic structure (FST = 0.11). The best partition for number of populations, based on 2nd order rate of change of the likelihood distribution, was K = 10 genetic clusters. The spatially explicit Bayesian clustering algorithm produced similar results, with minor differences in designation of admixed sites. We found evidence of past (and possibly ongoing) translocations; many populations were admixed. Our original goal was to identify landscape features, such as barriers or dispersal corridors, that could be used to aid management. Unfortunately, the extensive admixture among clusters made this impossible. This research shows that large-scale management of wild pigs may be necessary to achieve control and ameliorate damages. Reduction or cessation of translocations is necessary to prevent human-mediated dispersion of wild pigs.

Stochastic multi-step saddle point process minimax control parametric synthesis

The actual research problem of operations is development of methods of increase of a management efficiency by processes of the conflict nature. Article is devoted to development of methods of increase of a management efficiency by such processes. The purpose of article is the substantiation of the approach to parametrical synthesis of optimum control by multi-step stochastic minimax processes and procedures of the numerical analysis of likelihood dynamic characteristics of process. Formalization of process consists in definition of its type, a vector of phase coordinates and corresponding restrictions, the task of set of the actions sold by each the parties, efficiency of each action, control parameters (varied parameters of process) which task of values each of the parties influences a course of process, control restrictions, criteria of efficiency of the parties expressed through elements of a vector of phase coordinates. Discrete final stochastic process is considered. Change of phase coordinates occurs during the discrete moments of time, named steps of process. Phase coordinates depend on values of two groups of control parameters (controls of the counteracting parties). Within the limits of the modern theory of optimization of stochastic systems procedure of synthesis of optimum control is realized two-phase. At the first stage with use of analytical methods the structure of optimum control is determined. For these purposes the simplified determined model of process can to be used. At the second stage parametrical control optimization with use of algorithmic methods and computing procedures statistical linearization is carried out. Dynamics of process is described vectorial finite-difference equation. It is necessary to distinguish cases when there is saddle a point and when saddle the point is absent. Parametrical synthesis of optimum control is possible only in the first case. It is considered three basic variants of the equation: the linear equation; the nonlinear equation with optimum controls on border of a range of definition; the nonlinear equation with optimum controls inside of a range of definition. For the first variant there is an effective algorithm of parametrical synthesis of optimum control. For the second variant of synthesis of optimum control it is possible, but the algorithm is not effective. For the third variant to determine optimum managements it is not possible. Procedure statistical linearization is offered. Procedure consists in generation of set of realizations of the casual process set by the vector equation, calculation of optimum control for each concrete realization and the further statistical processing of the received results. The process described by the piecewise linear vector equation, is a special case of nonlinear process. At that it keeps property of independence of optimum control from coordinates of process. It provides expansion of a scope of effective computing procedure of synthesis of optimum control on a new class of piecewise linear processes. Property of a constancy process Hamiltonian can be used as criterion of correctness of calculation of optimum control in concrete cases. Application of the offered procedure provides use of methods of statistical modelling for the decision of tasks of the analysis of dynamics of the conflict and synthesis of optimum control in view of nonlinearity of functions of losses of the parties, dependence of efficiency of means used by them on the random factors formalized in the form of stochastic functions with various likelihood distributions, and also uncertainty concerning actions of the opponent.

Likelihood Estimation and Wavelet Transformation Based Optimization for Minimization of Noisy Pixels

In recent times the statistical computation techniques are gaining a lot of interest for analyzing the behavior of various mathematical distributions. This paper derives the likelihood distribution of image priors which has been further used to denoise the image. This paper aims to maximize the likelihood estimation of the parameters of interest i.e. prior and posterior estimation. We have proposed a prior-based distribution model which has been applied to additive, multiplicative and mixed noise cases. The various estimation parameters such as statistical variance and mean parameters have been used to evaluate the maximum likelihood of image priors for these noise models. Later, we have used an optimization technique based on the likelihood to reconstruct noise-free images efficiently. This paper uses conditional likelihood and wavelet transformation-based minimization techniques to minimize the noise in the pixels and a final denoised image is recovered. The conditional likelihood of the image has been optimized using pixel-based minimization w.r.t. the wavelet transformation coefficients. The simulation and analytical results have also been presented for the different noise cases.

An Optimal Higher Order Likelihood Distribution Based Approach for Strong Edge and High Contrast Restoration

The existing studies are generally focused on the pixel intensity or noise variance, and not much attention is given to describe the actual distribution. Some modifications have been proposed over the years in conventional parametric estimation techniques such as maximum likelihood estimation (MLE) and Bayesian estimation (BE). These methods have better properties to provide visually pleasing results for any imagery. In this paper, we use the likelihood estimate from the perspective that as the distribution order of likelihood increases, the overall effect of parameters under consideration improves within the defined confidence interval. In this paper, we use two parameters of interest, i.e., contrast and edge information. The proposed idea uses prior information w.r.t. particular parameter of interest, which is derived from likelihood estimate. The prior estimation has also been used to compute the reliability of the point estimations under a confidence interval. The proposed idea has desirable properties such that its optimization improves the overall image reconstruction.

Multifractal point processes and the spatial distribution of wildfires in French Mediterranean regions

We introduce a simple and wide class of multifractal spatial point patterns as Cox processes which intensity is multifractal, i.e., the class of Poisson processes with a stochastic intensity corresponding to a random multifractal measure. We then propose a maximum likelihood approach by means of a standard Expectation–Maximization procedure in order to estimate the distribution of these intensities at all scales. This provides, as validated on various numerical examples, a simple framework to estimate the scaling laws and therefore the multifractal properties for this class of spatial point processes. The wildfire distribution gathered in the Prométhée French Mediterranean wildfire database is investigated within this approach that notably allows us to compute the statistical moments associated with the spatial distribution of annual likelihood of fire event occurrence. We show that for each order q, these moments display a well defined scaling behavior with a non-linear spectrum of scaling exponents ζq. From our study, it thus appears that the spatial distribution of the wildfire ignition annual risk can be described by a non-trivial, multifractal singularity spectrum and that this risk cannot be reduced to providing a number of events per km2. Our analysis is confirmed by a direct spatial correlation estimation of the intensity logarithms whose the peculiar slowly decreasing shape corresponds to the hallmark of multifractal cascades. The multifractal features appear to be constant over time and similar over the three regions that are studied.

Reliable fusion of ToF and stereo data based on joint depth filter

To obtain reliable depth images with high resolution, a novel method is proposed in this study that fuses data acquired from time-of-flight (ToF) and stereo cameras, through which the advantages of both active and passive sensing are utilised. Based on the classic error model of the ToF, gradient information is introduced to establish the likelihood distribution for all disparity candidates. The stereo likelihood is estimated in parallel based on a 3D adaptive support-weight approach. The two independent likelihoods are unified using a maximum likelihood estimation, a process also referred to as a joint depth filter herein. Conventional post-processing methods such as a mutual consistency check are also used after applying a joint depth filter. We also propose a novel hole-filling method based on the seed-growing algorithm to retrieve missing disparities. Experiment results show that the proposed fusion method can produce reliable high-resolution depth maps and outperforms other compared methods.

A Sample Weight and AdaBoost CNN-Based Coarse to Fine Classification of Fruit and Vegetables at a Supermarket Self-Checkout

The physical features of fruit and vegetables make the task of vision-based classification of fruit and vegetables challenging. The classification of fruit and vegetables at a supermarket self-checkout poses even more challenges due to variable lighting conditions and human factors arising from customer interactions with the system along with the challenges associated with the colour, texture, shape, and size of a fruit or vegetable. Considering this complex application, we have proposed a progressive coarse to fine classification technique to classify fruit and vegetables at supermarket checkouts. The image and weight of fruit and vegetables have been obtained using a prototype designed to simulate the supermarket environment, including the lighting conditions. The weight information is used to change the coarse classification of 15 classes down to three, which are further used in AdaBoost-based Convolutional Neural Network (CNN) optimisation for fine classification. The training samples for each coarse class are weighted based on AdaBoost optimisation, which are updated on each iteration of a training phase. Multi-class likelihood distribution obtained by the fine classification stage is used to estimate a final classification with a softmax classifier. GoogleNet, MobileNet, and a custom CNN have been used for AdaBoost optimisation, with promising classification results.

A framework for modeling epistatic interaction.

Recently, various tools for detecting single nucleotide polymorphisms (SNPs) involved in epistasis have been developed. However, no studies evaluate the employed statistical epistasis models such as the χ2-test or quadratic regression independently of the tools that use them. Such an independent evaluation is crucial for developing improved epistasis detection tools, for it allows to decide if a tool's performance should be attributed to the epistasis model or to the optimization strategy run on top of it. We present a protocol for evaluating epistasis models independently of the tools they are used in and generalize existing models designed for dichotomous phenotypes to the categorical and quantitative case. In addition, we propose a new model which scores candidate SNP sets by computing maximum likelihood distributions for the observed phenotypes in the cells of their penetrance tables. Extensive experiments show that the proposed maximum likelihood model outperforms three widely used epistasis models in most cases. The experiments also provide valuable insights into the properties of existing models, for instance, that quadratic regression perform particularly well on instances with quantitative phenotypes. The evaluation protocol and all compared models are implemented in C++ and are supported under Linux and macOS. They are available at https://github.com/baumbachlab/genepiseeker/, along with test datasets and scripts to reproduce the experiments. Supplementary data are available at Bioinformatics online.

WHAT THE FUTURE MIGHT HOLD: DISTRIBUTIONS OF REGIONAL SECTORAL DAMAGES FOR THE UNITED STATES — ESTIMATES AND MAPS IN AN EXHIBITION

The text and associated Supplemental Materials contribute internally consistent and therefore entirely comparable regional, temporal, and sectoral risk profiles to a growing literature on regional economic vulnerability to climate change. A large collection of maps populated with graphs of Monte-Carlo simulation results support a communication device in this regard — a convenient visual that we hope will make comparative results tractable and credible and resource allocation decisions more transparent. Since responding to climate change is a risk-management problem, it is important to note that these results address both sides of the risk calculation. They characterize likelihood distributions along four alternative emissions futures (thereby reflecting the mitigation side context); and they characterize consequences along these transient trajectories (which can thereby inform planning for the iterative adaptation side). Looking across the abundance of sectors that are potentially vulnerable to some of the manifestations of climate change, the maps therefore hold the potential of providing comparative information about the magnitude, timing, and regional location of relative risks. This is exactly the information that planners who work to protect property and public welfare by allocating scarce resources across competing venues need to have at their disposal — information about relative vulnerabilities across time and space and contingent on future emissions and future mitigation. It is also the type of information that integrated assessment researchers need to calibrate and update their modeling efforts — scholars who are exemplified by Professor Nordhaus who created and exercised the Dynamic Integrated Climate-Economy and Regional Integrated Climate-Economy models.