Bayesian Selection Research Articles

Bayesian evidence-driven likelihood selection for sky-averaged 21-cm signal extraction

AbstractWe demonstrate that the Bayesian evidence can be used to find a good approximation of the ground truth likelihood function of a dataset, a goal of the likelihood-free inference (LFI) paradigm. As a concrete example, we use forward modelled sky-averaged 21-cm signal antenna temperature datasets where we artificially inject noise structures of various physically motivated forms. We find that the Gaussian likelihood performs poorly when the noise distribution deviates from the Gaussian case, for example, heteroscedastic radiometric or heavy-tailed noise. For these non-Gaussian noise structures, we show that the generalised normal likelihood is on a similar Bayesian evidence scale with comparable sky-averaged 21-cm signal recovery as the ground truth likelihood function of our injected noise. We therefore propose the generalised normal likelihood function as a good approximation of the true likelihood function if the noise structure is a priori unknown.

A Novel Nonconvex Low-Rank Tensor Completion Approach for Traffic Sensor Data Recovery From Incomplete Measurements

Complete traffic sensor data is considered to be one of the critical ingredients for intelligent transportation systems (ITS). However, the traffic measurements prevalently suffer from the inevitable and ubiquitous missing values. Current missing data completion algorithms are difficult to leverage the global low-rank property and the fine-grained spatial-temporal structure simultaneously. To circumvent this problem, this work presents a novel collaborative nonconvex low-rank spatial-temporal data tensor completion model that can take full advantage of the inherent spatial-temporal characteristics of traffic measurement data. First, the tensor Schatten p-norm, as an effective nonconvex surrogate of rank function, is used to exploit the global multi-dimensional correlation of traffic data. Then, we present an elastic net self-representation method and utilize an autoregression model in order to simultaneously capture the self-similarity and the temporal continuity of traffic data acquired in the same sensor network. By integrating the above elements in a unified nonconvex learning model, our method can explore the inherent structure of traffic data from the viewpoints of both global multi-dimensional correlation and fine-grained spatial and temporal dependency. Then, in the view of the general framework of the alternating directions method of multipliers (ADMM), an efficient iterative algorithm is designed to solve our model. Besides, to optimize the parameter combination of the model, a Bayesian optimization-based parameter selection algorithm is developed, which avoids manual parameter adjustment. Extensive experiments and analyses on two real-world traffic datasets are carried out. The results demonstrate the advantages of our model under diverse missing patterns and missing ratios.

Spatiotemporal Evolution and Drivers of Total Health Expenditure across Mainland China in Recent Years.

A substantially growing health expenditure has become an important global issue. Thus, how and why health expenditure is rising should be urgently investigated in systematic research. The Bayesian space-time model and the Bayesian least absolute shrinkage and selection operator (LASSO) model were employed in this study to investigate the spatiotemporal trends and influence patterns of total health expenditure per capita (THEPC) and total health expenditure (THEE) as a share of the gross domestic product (GDP) on the Chinese mainland from 2009 to 2018. The spatial distribution of THEE as a share of GDP in mainland China has shaped a distinct geographical structure with the characteristic of 'west high/east low'. Its local increasing trends formed a geographical structure that exhibited a 'north high/south low' feature. The heterogeneity of the influence patterns of health expenditure was observed from east to west across China. Natural environmental factors, such as air pollution and green coverage, along with changes in dietary structures, have increasingly influenced the growth of health expenditures.

Inferring selection effects in SARS-CoV-2 with Bayesian Viral Allele Selection.

The global effort to sequence millions of SARS-CoV-2 genomes has provided an unprecedented view of viral evolution. Characterizing how selection acts on SARS-CoV-2 is critical to developing effective, long-lasting vaccines and other treatments, but the scale and complexity of genomic surveillance data make rigorous analysis challenging. To meet this challenge, we develop Bayesian Viral Allele Selection (BVAS), a principled and scalable probabilistic method for inferring the genetic determinants of differential viral fitness and the relative growth rates of viral lineages, including newly emergent lineages. After demonstrating the accuracy and efficacy of our method through simulation, we apply BVAS to 6.9 million SARS-CoV-2 genomes. We identify numerous mutations that increase fitness, including previously identified mutations in the SARS-CoV-2 Spike and Nucleocapsid proteins, as well as mutations in non-structural proteins whose contribution to fitness is less well characterized. In addition, we extend our baseline model to identify mutations whose fitness exhibits strong dependence on vaccination status as well as pairwise interaction effects, i.e. epistasis. Strikingly, both these analyses point to the pivotal role played by the N501 residue in the Spike protein. Our method, which couples Bayesian variable selection with a diffusion approximation in allele frequency space, lays a foundation for identifying fitness-associated mutations under the assumption that most alleles are neutral.

Bayesian Molecular Dating Analyses Combined with Mutational Profiling Suggest an Independent Origin and Evolution of SARS-CoV-2 Omicron BA.1 and BA.2 Sub-Lineages.

The ongoing evolution of severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) has resulted in the recent emergence of a highly divergent variant of concern (VOC) defined as Omicron or B.1.1.529. This VOC is of particular concern because it has the potential to evade most therapeutic antibodies and has undergone a sustained genetic evolution, resulting in the emergence of five distinct sub-lineages. However, the evolutionary dynamics of the initially identified Omicron BA.1 and BA.2 sub-lineages remain poorly understood. Herein, we combined Bayesian phylogenetic analysis, mutational profiling, and selection pressure analysis to track the virus's genetic changes that drive the early evolutionary dynamics of the Omicron. Based on the Omicron dataset chosen for the improved temporal signals and sampled globally between November 2021 and January 2022, the most recent common ancestor (tMRCA) and substitution rates for BA.1 were estimated to be that of 18 September 2021 (95% highest posterior density (HPD), 4 August-22 October 2021) and 1.435 × 10-3 (95% HPD = 1.021 × 10-3 - 1.869 × 10-3) substitution/site/year, respectively, whereas 3 November 2021 (95% highest posterior density (HPD) 26 September-28 November 2021) and 1.074 × 10-3 (95% HPD = 6.444 × 10-4 - 1.586 × 10-3) substitution/site/year were estimated for the BA.2 sub-lineage. The findings of this study suggest that the Omicron BA.1 and BA.2 sub-lineages originated independently and evolved over time. Furthermore, we identified multiple sites in the spike protein undergoing continued diversifying selection that may alter the neutralization profile of BA.1. This study sheds light on the ongoing global genomic surveillance and Bayesian molecular dating analyses to better understand the evolutionary dynamics of the virus and, as a result, mitigate the impact of emerging variants on public health.

Bayesian model and selection signature analyses reveal risk factors for canine atopic dermatitis.

Canine atopic dermatitis is an inflammatory skin disease with clinical similarities to human atopic dermatitis. Several dog breeds are at increased risk for developing this disease but previous genetic associations are poorly defined. To identify additional genetic risk factors for canine atopic dermatitis, we here apply a Bayesian mixture model adapted for mapping complex traits and a cross-population extended haplotype test to search for disease-associated loci and selective sweeps in four dog breeds at risk for atopic dermatitis. We define 15 associated loci and eight candidate regions under selection by comparing cases with controls. One associated locus is syntenic to the major genetic risk locus (Filaggrin locus) in human atopic dermatitis. One selection signal in common type Labrador retriever cases positions across the TBC1D1 gene (body weight) and one signal of selection in working type German shepherd controls overlaps the LRP1B gene (brain), near the KYNU gene (psoriasis). In conclusion, we identify candidate genes, including genes belonging to the same biological pathways across multiple loci, with potential relevance to the pathogenesis of canine atopic dermatitis. The results show genetic similarities between dog and human atopic dermatitis, and future across-species genetic comparisons are hereby further motivated.

Performance Analysis of Network Selection Using ELECTRE Method

In this type of networks, overall network stability, utilization of resources, User satisfaction and most conducive to quality of service and selecting a Trusted access network (AN) is network in HWN (ANS) paper Proposes plan proposed project user, Can consider operator and/or quos perspectives general multicriteria software assistant (SA) was used to present and design. In imperfect wireless networks Bayesian information on the network selection problem we make it a game. Generally, a mobile user's preference (i.e., usage) is personal information. Therefore, each user provides others an optimal network selection decision must be made. Bayesian Nash equilibrium is assumed to be the solution of this game, also between the Bayesian Nash equilibrium and the equilibrium distribution of aggregate dynamics there is a one-to-one mapping. This is a Bayesian network selection game Integration of overall optimal response dynamics Numerical results. Show that Referring to the notification, the current paper is called Exams presents a review on the application of MCDM to determine the best PC architecture among three PC architectures, i.e.: A, B&C using ELECTRE technique based on four criteria. Provides a dynamic response to the problem. Henceforth this strategy is incorporated in this study. Arrangements obtained by AHP and ELECTRE technique are considered in the present study. Alternatives (eg wireless networks) represent attributes such as Network Selection 1, Network Selection 2, Network Selection 3, Network Selection 4, and Network Selection 5. Delay (msec), Jitter (msec), BER (• 10), Throughput (kbps), Cost (units) is an evaluation method. In this type of analysis, Gray Correlation Analysis (GRA) methods determine the best solution for a short distance to negativity and a long distance to solution. As a result, Wireless Network 5 ranked first and Wireless Network 4 ranked lowest.

Abstract 13566: Predictors of Progressive QRS Duration in Left Bundle Branch Block

Introduction: QRS duration in patients with left bundle branch block (LBBB) predicts adverse clinical outcomes and response to cardiac resynchronization therapy (CRT). Changes in LBBB QRS duration over time have not been reported. Hypothesis: QRS duration in patients with LBBB changes over time and may be predicted by patient demographics and comorbidities. Methods: We retrospectively identified 450 patients with LBBB on an index ECG (QRS ≥120 msec) recorded between 1992 and 2000 and examined any follow-up ECGs recorded after approximately 1, 2, 3, 5, and 10 years. Piecewise linear interpolations were performed, and separate generalized linear models (GLM) with Schwartz Bayesian Criterion effect selection were used to relate baseline clinical variables to changes in QRS duration at follow-up. Results: The cohort was 53.1% male, average age 66.7 yr, and mean index ECG QRS duration 144.7 msec. Although there was significant variability, the duration of the mean QRS increased as a tapering exponential function of time (p<0.0001) with progressively less increase at later follow-up. Three additional variables strongly and independently predicted progressive lengthening of the QRS within the overall GLM: heart failure with reduced ejection fraction (HFrEF) (p<0.0001), index ECG QRS duration (neg) (p<0.0001), and male gender (p=0.0063). In a subgroup of patients with an index ECG QRS <150 msec, these four GLM model variables also predicted the increase in QRS: time (p=0.0129), index QRS duration (p=0.0042), HFrEF (p=0.0001), and male gender (p=0.0084). (Equation, Figure below) ΔQRS(t;i) = 50.22 - 9.71*e -0.29*t - 0.344*QRS(0;i) + 7.62* 1 HFrEF (i) +5.12* 1 Male (i) Adjusted R 2 = 0.103 Conclusions: QRS duration progressively prolongs with time in patients with LBBB especially in patients with relatively narrow QRS and can be predicted by the presence of HFrEF and male gender. These predictors may be useful in better informing clinical prognosis and decisions regarding CRT.

GestroNet: A Framework of Saliency Estimation and Optimal Deep Learning Features Based Gastrointestinal Diseases Detection and Classification.

In the last few years, artificial intelligence has shown a lot of promise in the medical domain for the diagnosis and classification of human infections. Several computerized techniques based on artificial intelligence (AI) have been introduced in the literature for gastrointestinal (GIT) diseases such as ulcer, bleeding, polyp, and a few others. Manual diagnosis of these infections is time consuming, expensive, and always requires an expert. As a result, computerized methods that can assist doctors as a second opinion in clinics are widely required. The key challenges of a computerized technique are accurate infected region segmentation because each infected region has a change of shape and location. Moreover, the inaccurate segmentation affects the accurate feature extraction that later impacts the classification accuracy. In this paper, we proposed an automated framework for GIT disease segmentation and classification based on deep saliency maps and Bayesian optimal deep learning feature selection. The proposed framework is made up of a few key steps, from preprocessing to classification. Original images are improved in the preprocessing step by employing a proposed contrast enhancement technique. In the following step, we proposed a deep saliency map for segmenting infected regions. The segmented regions are then used to train a pre-trained fine-tuned model called MobileNet-V2 using transfer learning. The fine-tuned model's hyperparameters were initialized using Bayesian optimization (BO). The average pooling layer is then used to extract features. However, several redundant features are discovered during the analysis phase and must be removed. As a result, we proposed a hybrid whale optimization algorithm for selecting the best features. Finally, the selected features are classified using an extreme learning machine classifier. The experiment was carried out on three datasets: Kvasir 1, Kvasir 2, and CUI Wah. The proposed framework achieved accuracy of 98.20, 98.02, and 99.61% on these three datasets, respectively. When compared to other methods, the proposed framework shows an improvement in accuracy.

A Measure-on-Graph-Valued Diffusion: A Particle System with Collisions and Its Applications

A diffusion-taking value in probability-measures on a graph with vertex set V, ∑i∈Vxiδi is studied. The masses on each vertex satisfy the stochastic differential equation of the form dxi=∑j∈N(i)xixjdBij on the simplex, where {Bij} are independent standard Brownian motions with skew symmetry, and N(i) is the neighbour of the vertex i. A dual Markov chain on integer partitions to the Markov semigroup associated with the diffusion is used to show that the support of an extremal stationary state of the adjoint semigroup is an independent set of the graph. We also investigate the diffusion with a linear drift, which gives a killing of the dual Markov chain on a finite integer lattice. The Markov chain is used to study the unique stationary state of the diffusion, which generalizes the Dirichlet distribution. Two applications of the diffusions are discussed: analysis of an algorithm to find an independent set of a graph, and a Bayesian graph selection based on computation of probability of a sample by using coupling from the past.

Utility‐based Bayesian personalized treatment selection for advanced breast cancer

Utility‐based Bayesian personalized treatment selection for advanced breast cancer

General Bayesian Loss Function Selection and the use of Improper Models

AbstractStatisticians often face the choice between using probability models or a paradigm defined by minimising a loss function. Both approaches are useful and, if the loss can be re-cast into a proper probability model, there are many tools to decide which model or loss is more appropriate for the observed data, in the sense of explaining the data's nature. However, when the loss leads to an improper model, there are no principled ways to guide this choice. We address this task by combining the Hyvärinen score, which naturally targets infinitesimal relative probabilities, and general Bayesian updating, which provides a unifying framework for inference on losses and models. Specifically we propose the ℋ-score, a general Bayesian selection criterion and prove that it consistently selects the (possibly improper) model closest to the data-generating truth in Fisher's divergence. We also prove that an associated ℋ-posterior consistently learns optimal hyper-parameters featuring in loss functions, including a challenging tempering parameter in generalised Bayesian inference. As salient examples, we consider robust regression and non-parametric density estimation where popular loss functions define improper models for the data and hence cannot be dealt with using standard model selection tools. These examples illustrate advantages in robustness-efficiency trade-offs and enable Bayesian inference for kernel density estimation, opening a new avenue for Bayesian non-parametrics.

Nonparametric Bayesian functional selection in 1-M matched case-crossover studies.

The matched case-crossover study has been used in many areas such as public health, biomedical, and epidemiological research for humans, animals, and other subjects with clustered binary outcomes. The control information for each stratum is based on the subject's exposure experience, and the stratifying variable is the individual subject. It is generally accepted that any effects associated with the matching covariates by stratum can be removed in the conditional logistic regression model. However, when there are numerous covariates, it is important to perform variable selection to study the functional association between the variables and the relative risk of diseases or clustered binary outcomes by simultaneously adjusting effect modifications. The methods for simultaneously evaluating effect modifications by matching covariates such as time, as well as performing automatic variable and functional selections under semiparametric model frameworks, are quite limited. In this article, we propose a unified Bayesian approach due to its ability to detect both parametric and nonparametric relationships between the predictors and the relative risk of diseases or binary outcomes, accounting for potential effect modifications by matching covariates such as time, and perform automatic variable and functional selections. We demonstrate the advantages of our approach using simulation study and an epidemiological example of a 1-4 bidirectional case-crossover study.

The necrosis- and ethylene-inducing peptide 1-like protein (NLP) gene family of the plant pathogen Corynespora cassiicola.

Effectors are secreted by plant-associated microorganisms to modify the host cell physiology. As effectors, the Necrosis- and Ethylene-inducing peptide 1-like proteins (NLPs) are involded in the early phases of plant infection and may trigger host immune responses. Corynespora cassiicola is a polyphagous plant pathogen that causes target spot on many agriculturally important crops. Using genome assembly, gene prediction, and proteome annotation tools, we retrieved 135 NLP-encoding genes from proteomes of 44 isolates. We explored the evolutionary history of NLPs using Bayesian phylogeny, gene genealogies, and selection analyses. We accessed the expression profiles of the NLP genes during the early phase of C. cassiicola-soybean interaction. Three NLP putative-effector genes (Cc_NLP1.1, Cc_NLP1.2A, and Cc_NLP1.2B) were maintained in the genomes of all isolates tested. An NLP putative-non-effector gene (Cc_NLP1.3) was found in three isolates that had been originally obtained from soybean. Putative-effector NLPs were under different selective constraints: Cc_NLP1.1 was under stronger selective pressure, while Cc_NLP1.2A was under a more relaxed constraint. Meanwhile, Cc_NLP1.2B likely evolved under either positive or balancing selection. Despite highly divergent, the putative-effector NLPs maintain conserved the residues necessary to trigger plant immune responses, suggesting they are potentially functional. Only the Cc_NLP1.1 putative-effector gene was significantly expressed at the early hours of soybean colonization, while Cc_NLP1.2A and Cc_NLP1.2B showed much lower levels of gene expression.

Utility-based Bayesian personalized treatment selection for advanced breast cancer.

A Bayesian method is proposed for personalized treatment selection in settings where data are available from a randomized clinical trial with two or more outcomes. The motivating application is a randomized trial that compared letrozole plus bevacizimab to letrozole alone as first-line therapy for hormone receptor positive advanced breast cancer. The combination treatment arm had larger median progression-free survival time, but also a higher rate of severe toxicities. This suggests that the risk-benefit trade-off between these two outcomes should play a central role in selecting each patient's treatment, particularly since older patients are less likely to tolerate severe toxicities. To quantify the desirability of each possible outcome combination for an individual patient, we elicited from breast cancer oncologists a utility function that varied with age. The utility was used as an explicit criterion for quantifying risk-benefit trade-offs when making personalized treatment selections. A Bayesian nonparametric multivariate regression model with a dependent Dirichlet process prior was fit to the trial data. Under the fitted model, a new patient's treatment can be selected based on the posterior predictive utility distribution. For the breast cancer trial dataset, the optimal treatment depends on the patient's age, with the combination preferable for patients 70 years or younger and the single agent preferable for patients older than 70.

Bayesian treatment screening and selection using subgroup-specific utilities of response and toxicity.

A Bayesian design is proposed for randomized phase II clinical trials that screen multiple experimental treatments compared to an active control based on ordinal categorical toxicity and response. The underlying model and design account for patient heterogeneity characterized by ordered prognostic subgroups. All decision criteria are subgroup specific, including interim rules for dropping unsafe or ineffective treatments, and criteria for selecting optimal treatments at the end of the trial. The design requires an elicited utility function of the two outcomes that varies with the subgroups. Final treatment selections are based on posterior mean utilities. The methodology is illustrated by a trial of targeted agents for metastatic renal cancer, which motivated the design methodology. In the context of this application, the design is evaluated by computer simulation, including comparison to three designs that conduct separate trials within subgroups, or conduct one trial while ignoring subgroups, or base treatment selection on estimated response rates while ignoring toxicity.

An Approximate Bayesian Selection of Statistical Model Based on Fuzzy Interval Data

An Approximate Bayesian Selection of Statistical Model Based on Fuzzy Interval Data

Bayesian framework selection for hyperspectral image denoising

Mixture noise reduction is a key pre-process that improves the hyperspectral images’ (HSI) quality and prepares them for the next processes. Mixture noise should be modeled as similar as possible to the real HSI noise model, which is a challenge for noise removal methods. These use noise model simplification assumptions that can take the method performance away from real scenarios. This paper adopts a new general model for the mixture noise which leads to one model selection framework. The optimization problem of the proposed method is formed using Bayesian risk in five different models and promoted to total variation regularized low-rank matrix factorization. The devised optimization problem is solved using augmented Lagrange multiplier and proximal gradient algorithms. Also, we compared the proposed method with other state-of-the-art methods to reduce the mixture noise of HSI, which is a combination of Gaussian (i.i.d or non-i.i.d), and sparse (e.g., stripe, deadline, impulse) noises. The results obtained on both real and synthetic HSI data sets show the proposed method’s performance superiority to other competing methods in both visual comparisons and quantitative evaluations.

Bayesian reciprocal bridge composite tobit quantile regression

A composite tobit quantile regression approach is proposed for Bayesian simultaneous covariate selection and estimation in the setting of left censored regression. The proposed approach uses prior distributions for the regression coefficients that are scale mixtures of inverse uniform priors on the coefficients and independent Gamma priors on their mixing parameters. The proposed method was illustrated using simulation examples. Results show that the proposed method performs very well compared to the existing method.

Prophet Inequality for Bipartite Matching: Merits of Being Simple and Nonadaptive

We consider the Bayesian online selection problem of a matching in bipartite graphs, that is, the weighted online matching problem where the edges arrive online and edge weights are generated from a known distribution. This setting corresponds to the intersection of two matroids in the work of Kleinberg and Weinberg [ 40 ] and Feldman et al. [ 27 ]. We study a simple class of nonadaptive policies that we call vertex-additive policies. A vertex-additive policy assigns static prices to every vertex in the graph and accepts only those edges whose weight exceeds the sum of the prices on the edge endpoints. We show that there exists a vertex-additive policy with the expected payoff of at least one-third of the prophet’s payoff and present a gradient descent algorithm that quickly converges to the desired vector of vertex prices. Our results improve on the adaptive online policies of Kleinberg and Weinberg and Feldman et al. for the intersection of two matroids in two ways: our policy is nonadaptive and has a better approximation guarantee of 3 instead of the previous guarantees of 5.82 in Kleinberg and Weinberg and 5.43 in Feldman et al. We give a complementary lower bound of 2.25 for any online algorithm in the bipartite matching setting. Funding: This work was supported by Fundamental Research Funds of the Central Universities in China, a Science and Technology Innovation 2030 major project [“New Generation of Artificial Intelligence,” Project 2018AAA0100903], the Shanghai Municipal Education Commission Innovation Program, and the Program for Innovative Research Team of the Shanghai University of Finance and Economics (IRTSHUFE).