Global Discrepancy Research Articles

Multi-Dimensional Visual Data Restoration: Uncovering the Global Discrepancy in Transformed High-Order Tensor Singular Values.

The recently proposed high-order tensor algebraic framework generalizes the tensor singular value decomposition (t-SVD) induced by the invertible linear transform from order-3 to order-d (d > 3). However, the derived order-d t-SVD rank essentially ignores the implicit global discrepancy in the quantity distribution of non-zero transformed high-order singular values across the higher modes of tensors. This oversight leads to sub-optimal restoration in processing real-world multi-dimensional visual datasets. To address this challenge, in this study, we look in-depth at the intrinsic properties of practical visual data tensors, and put our efforts into faithfully measuring their high-order low-rank nature. Technically, we first present a novel order-d tensor rank definition. This rank function effectively captures the aforementioned discrepancy property observed in real visual data tensors and is thus called the discrepant t-SVD rank. Subsequently, we introduce a nonconvex regularizer to facilitate the construction of the corresponding discrepant t-SVD rank minimization regime. The results show that the investigated low-rank approximation has the closed-form solution and avoids dilemmas caused by the previous convex optimization approach. Based on this new regime, we meticulously develop two models for typical restoration tasks: high-order tensor completion and high-order tensor robust principal component analysis. Numerical examples on order-4 hyperspectral videos, order-4 color videos, and order-5 light field images substantiate that our methods outperform state-of-the-art tensor-represented competitors. Finally, taking a fundamental order-3 hyperspectral tensor restoration task as an example, we further demonstrate the effectiveness of our new rank minimization regime for more practical applications. The source codes of the proposed methods are available at https://github.com/CX-He/DTSVD.git.

AnANet: Association and Alignment Network for Modeling Implicit Relevance in Cross-Modal Correlation Classification

With the explosive increase of multimodal data, cross-modal correlation classification has become an important research topic and is in great demand in many cross-modal applications. A variety of classification schemes and predictive models have been built based on the existing cross-modal correlation categorization. However, these classification schemes typically follow the prior assumption that the paired cross-modal samples are strictly related, and thus pay great attention to the fine-grained relevant types of cross-modal correlation, ignoring the high volume of implicitly relevant data which are often wrongly classified into irrelevant types. Even more, previous predictive models fall short of reflecting the essence of cross-modal correlation according to their definitions, especially in the modeling of network structure. Thus in this paper, by comprehensively investigating the current image-text correlation classification research, we redefine a new classification scheme for cross-modal correlation based on the implicit and explicit relevance. To predict the types of image-text correlation based on our proposed definition, we further devise the Association and Alignment Network (namely AnANet) to model the implicit and explicit relevance, which captures both the implicit association of global discrepancy and commonality between image and text and explicit alignment of cross-modal local relevance. Experimental studies on our constructed new image-text correlation dataset verify the effectiveness of our proposed model.

Factors associated with the Big 5 personality traits in Alzheimer’s disease and related dementias (ADRD)

AbstractBackgroundInformant ratings of behavior and personality change are the gold standard of measurement in Alzheimer’s disease and related dementias (ADRD) where insight is often impaired, but informant report can be biased by psychosocial factors such as caregiver burden. This study assessed the impact of caregiver burden on informant ratings of patient Big 5 personality traits (Extraversion, Agreeableness, Conscientiousness, Neuroticism, Openness) and associations between degree of discrepancy between informant and patient ratings with cortical volume in regions implicated in loss of insight.MethodA sample of 70 ADRD participants with heterogeneous neurodegenerative clinical phenotypes (Figure 1) and their informants completed the NEO Personality Inventory (NEO‐PI). Disease severity was measured using the Clinical Dementia Rating Scale (CDR). Caregiver burden was measured using the Zarit Burden Interview (ZBI). Discrepancy scores were computed for all NEO‐PI Big 5 scores and a global discrepancy score. Participants were classified as having high/low discrepancy using a median split for each NEO‐PI discrepancy score. T1‐weighted 3T MRI grey matter volumes were normalized to intracranial volume for all ROIs in the Schaefer100×17 parcellation, and linear regression analyses related global difference score to Bonferroni‐corrected atrophy regions.ResultHigher levels of caregiver burden were associated with higher informant ratings of patient Neuroticism (ß = 0.02, p = .013) and lower informant ratings of patient Agreeableness (ß = ‐0.02, p = .0004), Conscientiousness (ß = ‐0.02, p = .005), and Openness (ß = ‐0.01, p = .014) independent of disease severity. Patients with high global discrepancy scores showed reduced cortical volumes in left ventral PFC, right medial PFC, and right superior temporal gyrus. Patients with low discrepancy scores showed relatively preserved cortical volumes in bilateral somatomotor cortex, right posterior cingulate, and left supplementary motor, superior parietal lobule, and postcentral gyrus (all p‐values<.05, Figure 2). Reduced volumes in the high discrepancy group were driven by patients with bvFTD (Figure 3).ConclusionInformant ratings of Big 5 personality factors may be confounded by caregiver burden, highlighting the need for more objective measures of personality and behavior in ADRD. Additionally, high levels of discrepancy between informant and patient ratings of personality may reflect loss of insight secondary to cortical atrophy in frontal and temporal structures.

Abnormal-Aware Multiperson Evaluation System with Improved Fuzzy Weighting

There exists a phenomenon that subjectivity highly lies in the daily evaluation process. Our research primarily concentrates on a multiperson evaluation system with anomaly detection to minimize the possible inaccuracy that subjective assessment brings. We choose the two-stage screening method, which consists of rough screening and score-weighted Kendall-τ distance to winnow out abnormal data, coupled with hypothesis testing to narrow global discrepancy. Then we use fuzzy synthetic evaluation method (FSE) to determine the significance of scores given by reviewers as well as their reliability, culminating in a more impartial weight for each reviewer in the final conclusion. The results demonstrate a clear and comprehensive ranking instead of unilateral scores, and we get to have an efficiency in filtering out abnormal data as well as a reasonably objective weight determination mechanism. We can sense that through our study, people will have a chance of modifying a multiperson evaluation system to attain both equity and a relatively superior competitive atmosphere. A preprint has previously been published (Ni, 2022).

Adaptive parameter selection for weighted-TV image reconstruction problems

We propose an efficient estimation technique for the automatic selection of locally-adaptive Total Variation regularisation parameters based on an hybrid strategy which combines a local maximum-likelihood approach estimating space-variant image scales with a global discrepancy principle related to noise statistics. We verify the effectiveness of the proposed approach solving some exemplar image reconstruction problems and show its outperformance in comparison to state-of-the-art parameter estimation strategies, the former weighting locally the fit with the data [4], the latter relying on a bilevel learning paradigm [8, 9].

Does genome surveillance explain the global discrepancy between binding and effect of chromatin factors?

Knocking out a chromatin factor often does not alter the transcription of its binding targets. What explains the observed disconnect between binding and effect? We hypothesize that this discrepancy could be associated with the role of chromatin factors in maintaining genetic and epigenetic integrity at promoters, and not necessarily with transcription. Through re-analysis of published datasets, we present several lines of evidence that support our hypothesis and deflate the popular assumptions. We also tested the hypothesis through mutation accumulation assays on yeast knockouts of chromatin factors. Altogether, the proposed hypothesis presents a simple explanation for the global discord between chromatin factor binding and effect. Future work in this direction might fortify the hypothesis and elucidate the underlying mechanisms.

Regional analysis on model validation with different metrics

Compared with the global analysis, the regional analysis can offer more comprehensive and more detail information with an input variable fixed in an interval. Existing validation metrics mostly focus on the global discrepancy between the observations and the computer models. To further measure the difference in detail, the concept of regional validation was proposed in this work. To make regional validation specific and easily understandable, two existing validation metrics-the area metric and the G metric were applied. The contribution to whole validation (CWV) metric was proposed to depict the relationship between the proposed regional validation and the initial whole scale validation. At last, an example was employed to demonstrate the rationality and necessity of the regional validation.

EnsNet: Ensconce Text in the Wild

A new method is proposed for removing text from natural images. The challenge is to first accurately localize text on the stroke-level and then replace it with a visually plausible background. Unlike previous methods that require image patches to erase scene text, our method, namely ensconce network (EnsNet), can operate end-to-end on a single image without any prior knowledge. The overall structure is an end-to-end trainable FCN-ResNet-18 network with a conditional generative adversarial network (cGAN). The feature of the former is first enhanced by a novel lateral connection structure and then refined by four carefully designed losses: multiscale regression loss and content loss, which capture the global discrepancy of different level features; texture loss and total variation loss, which primarily target filling the text region and preserving the reality of the background. The latter is a novel local-sensitive GAN, which attentively assesses the local consistency of the text erased regions. Both qualitative and quantitative sensitivity experiments on synthetic images and the ICDAR 2013 dataset demonstrate that each component of the EnsNet is essential to achieve a good performance. Moreover, our EnsNet can significantly outperform previous state-of-the-art methods in terms of all metrics. In addition, a qualitative experiment conducted on the SBMNet dataset further demonstrates that the proposed method can also preform well on general object (such as pedestrians) removal tasks. EnsNet is extremely fast, which can preform at 333 fps on an i5-8600 CPU device.

Niche conservatism drives a global discrepancy in palm species richness between seasonally dry and moist habitats

AbstractAimRapid global environmental change predicts increasingly seasonal climate in the tropics, causing expansion of seasonally dry habitats and leading to shifts in species distribution and potential extinction. Here, we use a macroevolutionary framework to understand the processes driving palm diversity patterns between moist and seasonally dry tropical habitats. We hypothesize that the discrepancy in species richness between habitats is explained by higher speciation rates in moist habitat and that niche conservatism prevents frequent shifts between habitats.LocationGlobal.Time periodLast 100 Ma.Major taxa studiedArecaceae.MethodsWe used trait‐dependent diversification models to test whether different habitats affect palm speciation rates. Furthermore, palm assemblages were divided into three regions (Africa, Australasia and Neotropics) to test for niche conservatism and evaluate phylogenetic dissimilarity.ResultsWe found no relationship between speciation rate and habitat type. We detected phylogenetic signal for habitats at both global and continental scales, indicating that closely related species were more similar than expected by chance. Colonization of seasonally dry habitats occurred c. 60 Ma, yet most clades only diversified after c. 30 Ma. The high phylogenetic dissimilarity between habitat types at both global and continental scales was driven by high lineage turnover, at least for Africa and the Neotropics.Main conclusionsWe found a lack of differential speciation rates in setting the seasonally dry and moist palm‐richness discrepancy. However, over evolutionary history most palm lineages fail to colonize seasonally dry habitats owing to a tendency to retain ancestral habitat. Indeed, seasonally dry palm assemblages are the result of the diversification of particular lineages. Likewise, long‐term dry periods appear to induce shifts in taxonomic and functional diversity, and we emphasize that the expansion of dry habitats might also imply a loss of palm clades, hence a reduction in phylogenetic diversity.

What is the true tuberculosis mortality burden? Differences in estimates by the World Health Organization and the Global Burden of Disease study

The World Health Organization (WHO) and the Global Burden of Disease (GBD) study at the Institute for Health Metrics and Evaluation (IHME) periodically provide global estimates of tuberculosis (TB) mortality. We compared the 2015 WHO and GBD TB mortality estimates and explored which factors might drive the differences. We extracted the number of estimated TB-attributable deaths, disaggregated by age, HIV status, sex and country from publicly available WHO and GBD datasets for the year 2015. We 'standardized' differences between sources by adjusting each country's difference in absolute number of deaths by the average number of deaths estimated by both sources. For 195 countries with estimates from both institutions, WHO estimated 1768482 deaths attributable to TB, whereas GBD estimated 1322916 deaths, a difference of 445566 deaths or 29% of the average of the two estimates. The countries with the largest absolute differences in deaths were Nigeria (216621), Bangladesh (49863) and Tanzania (38272). The standardized difference was not associated with HIV prevalence, prevalence of multidrug resistance or global region, but did show correlation with the case detection rate as estimated by WHO [r = -0.37, 95% confidence interval (CI): -049; -0.24] or, inversely, with case detection rate based on GBD data (r = 0.44, 95% CI: 0.31; 0.54). Countries with a recent national prevalence survey had higher standardized differences (higher estimates by WHO) than those without (P = 0.006). After exclusion of countries with recent prevalence surveys, the overall correlation between both estimates was r = 0.991. A few countries account for the large global discrepancy in TB mortality estimates. The differences are due to the methodological approaches used by WHO and GBD. The use and interpretation of prevalence survey data and case detection rates seem to play a role in the observed differences.

Minimising disparity in distribution for unsupervised domain adaptation by preserving the local spatial arrangement of data

Domain adaptation is used for machine learning tasks, when the distribution of the training (obtained from source domain) set differs from that of the testing (referred as target domain) set. In the work presented in this study, the problem of unsupervised domain adaptation is solved using a novel optimisation function to minimise the global and local discrepancies between the transformed source and the target domains. The dissimilarity in data distributions is the major contributor to the global discrepancy between the two domains. The authors propose two techniques to preserve the local structural information of source domain: (i) identify closest pair of instances in source domain and minimise the distances between these pairs of instances after transformation; (ii) preserve the naturally occurring clusters present in source domain during transformation. This cost function and constraints yield a non-linear optimisation problem, used to estimate the weight matrix. An iterative framework solves the optimisation problem, providing a sub-optimal solution. Next, using orthogonality constraint, an optimisation task is formulated in the Stiefel manifold. Performance analysis using real-world datasets show that the proposed methods perform better than a few recently published state-of-the-art methods.

Ultra-diffuse cluster galaxies as key to the MOND cluster conundrum

MOND reduces greatly the mass discrepancy in clusters of galaxies, but does leave a consistent global discrepancy of about a factor of two. It has been proposed, within the minimalist and purist MOND, that clusters harbor some indigenous, yet-undetected, cluster baryonic (dark) matter (CBDM). Its total amount has to be comparable with that of the observed hot gas. Following an initial discovery by van Dokkum & al. (2015a), Koda & al. (2015) have recently identified more than a thousand ultra-diffuse galaxy-like objects (UDGs) in the Coma cluster. Robustness of the UDGs to tidal disruption seems to require, within Newtonian dynamics, that they are much more massive than their observed stellar component. Here, I propound that a considerable fraction of the CBDM is internal to UDGs, which endows them with robustness. The rest of the CBDM objects formed in now-disrupted kin of the UDGs, and is dispersed in the intracluster medium. While the discovery of cluster UDGs is not in itself a resolution of the MOND cluster conundrum, it lends greater qualitative plausibility to CBDM as its resolution, for reasons I discuss. Alternatively, if the UDGs are only now falling into Coma, their large size and very low surface brightness could result from the adiabatic inflation due to the MOND external-field effect, as described in Brada & Milgrom (2000). I also consider briefly solutions to the conundrum that invoke more elaborate extensions of purist MOND, e.g., that in clusters, the MOND constant takes up larger-than-canonical values of the MOND constant.

Bayesian sensitivity analysis with the Fisher–Rao metric

We propose a geometric framework to assess sensitivity of Bayesian procedures to modelling assumptions based on the nonparametric Fisher–Rao metric. While the framework is general, the focus of this article is on assessing local and global robustness in Bayesian procedures with respect to perturbations of the likelihood and prior, and on the identification of influential observations. The approach is based on a square-root representation of densities, which enables analytical computation of geodesic paths and distances, facilitating the definition of naturally calibrated local and global discrepancy measures. An important feature of our approach is the definition of a geometric ϵ-contamination class of sampling distributions and priors via intrinsic analysis on the space of probability density functions. We demonstrate the applicability of our framework to generalized mixed-effects models and to directional and shape data.

A novel model validation methodology using synchrophasor measurements

The validation of simulation models is a crucial step in power system security analysis. In this paper, a new methodology is proposed to approach this problem, using events recorded by a synchrophasor measurement system. The method is based on the characteristics of a new discrepancy index, denoted here as Global Discrepancy Indicator (GDI), and trajectories correlation. The GDI gives a quantitative measure of the model discrepancy with respect to the real system. It allows for the integration of the three main parts of the validation process and permits the selection of the differences minimization interval in the calibration phase. In addition, a ranking of the GDI sensitivities is used to identify an initial subset of problematic model parameters. The correlations between trajectory differences and trajectory sensitivities aid in the selection of the final subset of candidates to problematic parameters for calibration. The extended Least Squares Estimation is then used to automatically tune the identified problematic parameters. The complete methodology was tested and has shown its capability to improve the models coherency.

Adherence to antiretroviral therapy in adolescents living with HIV: systematic review and meta-analysis.

Objective:Adolescent and young adult (AYA) populations (12–24 years) represent over 40% of new HIV infections globally. Adolescence is sometimes characterized by high-risk sexual behaviour and a lack of engagement with healthcare services that can affect adherence to antiretroviral therapy (ART). Despite adherence to ART being critical in controlling viral replication, maintaining health and reducing onward viral transmission, there are limited data on ART adherence amongst AYA globally. We undertook a systematic review and meta-analysis of published studies reporting adherence to ART for AYA living with HIV.Design and methods:Searches included Embase, Medline and PsychINFO databases up to 14 August 2013. Eligible studies defined adequate adherence as at least 85% on self-report or undetectable blood plasma virus levels. A random effects meta-analysis was performed and heterogeneity examined using meta-regression.Results:We identified 50 eligible articles reporting data from 53 countries and 10 725 patients. Using a pooled analysis of all eligible studies, 62.3% [95% confidence interval (CI) 57.1–67.6; I2 : 97.2%] of the AYA population were adherent to therapy. The lowest average ART adherence was in North America [53% (95% CI 46–59; I2 : 91%)], Europe [62% (95% CI 51–73; I2 : 97%)] and South America [63% (95% CI 47–77; I2 : 85%] and, with higher levels in Africa [84% (95% CI 79–89; I2 : 93%)] and Asia [84% (95% CI 77–91; I2 : 0%].Conclusion:Review of published literature from Africa and Asia indicate more than 70% of HIV-positive AYA populations receiving ART are adherent to therapy and lower rates of adherence were shown in Europe and North America at 50–60%. The global discrepancy is probably multifactorial reflecting differences between focused and generalised epidemics, access to healthcare and funding.

A Local Weighted Mean Based Domain Adaptation Learning Framework

Maximum mean discrepancy(MMD),as a criterion effectively and effciently measuring the distribution discrepancy between source domains and target ones,has been successfully used.But it is a global measuring algorithm and to some extent only reflects the global distribution discrepancy between domains and the global structural difference.Therefore,we propose projected maximum local weighted mean discrepancy(PMLWD) scheme by with locality preserving ability integrating the theory and method of local weighted mean into the MMD.At the same time,we formulate in theory that the PMLWD is one of generalized algorithms of the MMD.Furthermore,on the basis of the PMLWD and by integrating classical learning theories,we present local weighted mean based domain adaptation learning framework(LDAF).Following the LDAF,we propose local weighted mean based multi-label classification domain adaptation learning algorithm(LDAF MLC) and local weighted mean based domain adaptation supporting vector machine(LDAF SVM).At last,tests on artificial data sets,high dimensional text data sets and face data sets show the LDAF methods are superior to other domain adaption ones.

Extensions of the skew-normal ogive item response model

We introduce new applications of the skew-probit IRT model by considering a flexible skew-normal distribution for the latent variables and by extending this model to include an additional random effect for modeling dependence between items within the same testlet. A Bayesian hierarchical structure is presented using a double data augmentation approach. This can be easily implemented in WinBUGS or SAS by considering MCMC algorithms. Several Bayesian model selection criteria, such as DIC, EAIC and EBIC, have been considered; in addition, we use posterior sum of squares of the latent residuals as a global discrepancy measure to model comparison. Two applications illustrate the methodology, one data set related to a mathematical test and another related to reading comprehension, both applied to Peruvian students. Results indicate better performance of the more flexible models proposed in this paper.

OP082: Effect of the tissue shrinkage phenomenon on surgical margins of resection in patients undergoing cancer oral and oropharynx

Purpose To evaluate the phenomenon of tissue shrinkage in surgical specimens of patients which underwent surgery for cancer of the oral cavity and oropharynx and were reconstructed by means of microsurgical techniques. Methods A prospective, descriptive and observational study was carried out. All patients fulfilled inclusions criteria. Following variables were analyzed: age, sex, toxic habits, symptoms, histological study according to the protocol of Cancer of Head and Neck from Pathology Department of our hospital (tumor morphology, location, degree of tumor differentiation, TNM, pattern of invasion tumor, etc.), pre, intra and post-operatory surgical margin status, tumor markers, etc. Images of the surgical specimens and resection margin were taken. Results Forty-four per cent of the patients studied were male. The average time between the measurements in fresh in the operating room until fresh measurement in pathology department was about 20 min. There was a global discrepancy in measurements made in fresh after formalin fixation and measurements made by microscopic analysis, showing an overall decrease in the distance of the surgical margins of resection. Conclusions The status of surgical margins of resection has been considered a potential indicator of surgical quality, local recurrence and overall survival. The effect of tissue shrinkage is one aspect to consider in evaluating of resection margins. Its results entail important clinical implications since if margins are affected or nearby, the adjuvant radiotherapy necessary to complete the treatment could add morbidity in the management of these patients.

In Search of Lost Capital: An Estimation of Undeclared Portfolio Assets

The analyses of the international investment position and balance of payments statistics suggest that foreign assets held abroad are greatly underestimated, in particular in the sector of portfolio investments. The aim of this work is to test this hypothesis and to estimate the magnitude of under-reported assets. The approach is based on the comparison of mirror statistics on portfolio assets and liabilities, mainly using data coming from the Coordinated Portfolio Investment Survey (CPIS) conducted by the IMF, with the addition of information derived from several international databases. For the years from 2001 to 2010 the global discrepancy is estimated to be equal to 7.3% of world GDP on average. Different criteria have been adopted to attribute the share of the estimated under-reporting, particularly significant in the case of mutual funds issued by Luxembourg and the main off-shore centres, to the main euro area countries. Results vary from 6 to 10 per cent of national GDPs. If these amounts were added to national data, statistical consistency in international statistics would improve.

The external wealth of nations mark II: Revised and extended estimates of foreign assets and liabilities, 1970–2004

We construct estimates of external assets and liabilities for 145 countries for 1970–2004. We describe our estimation methods and key features of the data at the country and global level. We focus on trends in net and gross external positions, and the composition of international portfolios. We document the increasing importance of equity financing and the improvement in the external position for emerging markets, and the differing pace of financial integration between advanced and developing economies. We also show the existence of a global discrepancy between estimated foreign assets and liabilities, and identify the asset categories accounting for this discrepancy.