Multiple Inference Research Articles

Quality Monitoring and Assessment of Deployed Deep Learning Models for Network AIOps

Artificial intelligence (AI) has recently attracted a lot of attention, transitioning from research labs to a wide range of successful deployments in many fields, which is particularly true for deep learning (DL) techniques. Ultimately, DL models, being software artifacts, need to be regularly maintained and updated: AIOps is the logical extension of the DevOps software development practices to AI software applied to network operation and management. In the life cycle of a DL model deployment, it is important to assess the quality of deployed models, to detect “stale” models and prioritize their update. In this article, we cover the issue in the context of network management, proposing simple but effective techniques for quality assessment of individual inference, and for overall model quality tracking over multiple inferences, that we apply to two use cases, representative of the network management and image recognition fields.

Affect, Support, and Personal Factors: Multimodal Causal Models of One-on-one Coaching

Human one-on-one coaching involves complex multimodal interactions. Successful coaching requires teachers to closely monitor students' cognitive-affective states and provide support of optimal type, timing, and amount. However, most of the existing human tutoring studies focus primarily on verbal interactions and have yet to incorporate the rich aspects of multimodal cognitive-affective experiences. Meanwhile, the research community lacks principled methods to fully exploit complex multimodal data to uncover the causal relationships between coaching supports, students' cognitive-affective experiences, and their stable individual factors. We explore an analytical framework that is explainable and amenable to incorporating domain knowledge. The proposed framework combines statistical approaches in Sparse Multiple Canonical Correlation, causal discovery, and inference methods for observations. We demonstrate this framework using a multimodal one-on-one math problem-solving coaching dataset collected in naturalistic home environments involving parents and young children. The insights derived from our analyses may inform the design of effective technology-inspired interventions that are personalized and adaptive.

Improved group sequential Holm procedures for testing multiple correlated hypotheses over time

ABSTRACT Clinical trials can typically feature two different types of multiple inference: testing of more than one null hypothesis and testing at multiple time points. These modes of multiplicity are closely related mathematically but distinct statistically and philosophically. Regulatory agencies require strong control of the family-wise error rate (FWER), the risk of falsely rejecting any null hypothesis at any analysis. The correlations between test statistics at interim analyses and the final analysis are therefore routinely used in group sequential designs to achieve less conservative critical values. However, the same type of correlations between different comparisons, endpoints or sub-populations are less commonly used. As a result, FWER is in practice often controlled conservatively for commonly applied procedures. Repeated testing of the same null hypothesis may give changing results, when the hypothesis is rejected at an interim but accepted at the final analysis. The mathematically correct overall rejection is at odds with an inference theoretic approach and with common sense. We discuss these two issues, of incorporating correlations and how to interpret time-changing conclusions, and provide case studies where power can be increased while adhering to sound statistical principles.

CorDiffViz: an R package for visualizing multi-omics differential correlation networks

BackgroundDifferential correlation networks are increasingly used to delineate changes in interactions among biomolecules. They characterize differences between omics networks under two different conditions, and can be used to delineate mechanisms of disease initiation and progression.ResultsWe present a new R package, CorDiffViz, that facilitates the estimation and visualization of differential correlation networks using multiple correlation measures and inference methods. The software is implemented in R, HTML and Javascript, and is available at https://github.com/sqyu/CorDiffViz. Visualization has been tested for the Chrome and Firefox web browsers. A demo is available at https://diffcornet.github.io/CorDiffViz/demo.html.ConclusionsOur software offers considerable flexibility by allowing the user to interact with the visualization and choose from different estimation methods and visualizations. It also allows the user to easily toggle between correlation networks for samples under one condition and differential correlations between samples under two conditions. Moreover, the software facilitates integrative analysis of cross-correlation networks between two omics data sets.

Spontaneous state inferences.

Perceivers routinely draw inferences about others from their behavior in an attempt to make sense of the world. Previous research has established that spontaneous inferences include stable characteristics such as traits and a number of variable person-related concepts such as goals, intentions, and motivations. The current research investigated the occurrence of more general spontaneous state inferences. In a series of four preregistered studies (N = 883), we adapted two established experimental paradigms frequently used in spontaneous social inference research to the investigation of spontaneous trait and state inferences. In Studies 1 and 2, we observed evidence for the occurrence of spontaneous state inferences from state-implying statements. In Studies 3 and 4, we observed the simultaneous occurrence of spontaneous trait and state inferences from statements that allowed for both inferences. In a fifth study (N = 97), we provide evidence that people represent states and traits as functionally different: Participants judged the likelihood of behavioral repetition higher when the same behavior was related to a trait-inference than a state-inference. The observation of multiple simultaneous spontaneous inferences in the current research suggests that further theory building regarding the underlying mechanisms and processes of spontaneous impression formation in person perception from behavior is warranted. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

An impact of lipid profile and lipid lowering drugs on \u226570\u2009year olds of an upper socioeconomic class: a retrospective cohort study

BackgroundLife expectancy has greatly increased, generating an improvement in screening programs for disease prevention, lifesaving drugs and medical devices. The impact of lowering low-density lipoprotein cholesterol (LDL-C) in the very elderly is not well-established. Our aim was to explore the association of LDL-C, high density lipoprotein cholesterol (HDL-C) and lipid lowering drugs (LLDs) on cognitive decline, malignancies and overall survival.MethodsThis was a retrospective cohort study. Our study comprised 1498 (72.7%) males and 561 (27.3%) females, aged ≥70 who had attended the Institute for Medical Screening (IMS), Sheba Medical Center, Israel at least twice during 2013–2019. Data were obtained from the computerized database of the IMS. A manual quality control to identify potential discrepancies was performed.ResultsOverall, 6.3% of the subjects treated with LLDs (95/1421) versus 4.2% not treated (28/638), cognitively declined during the study years. No statistically significant effects of LDL-C, HDL-C and LLDs on cognitive decline were observed after correcting for age, prior stroke and other vascular risk factors. With regard to cancer, after adjusting for confounders and multiple inferences, no definite relationships were found.ConclusionsThis analysis of an elderly, high socioeconomic status cohort suggests several relationships between the use of LLDs and health outcomes, some beneficial, especially, with regard to certain types of cancer, but with a higher risk of cognitive decline. Further studies are warranted to clarify the health effects of these medications in the elderly.

The Effects of Rock Index Tests on Prediction of Tensile Strength of Granitic Samples: A Neuro-Fuzzy Intelligent System

Rock tensile strength (TS) is an essential parameter for designing structures in rock-based projects such as tunnels, dams, and foundations. During the preliminary phase of geotechnical projects, rock TS can be determined through laboratory works, i.e., Brazilian tensile strength (BTS) test. However, this approach is often restricted by laborious and costly procedures. Hence, this study attempts to estimate the BTS values of rock by employing three non-destructive rock index tests. BTS predictive models were developed using 127 granitic rock samples. Since the simple regression analysis did not yield a meaningful result, the development of models that integrate multiple input parameters were considered to improve the prediction accuracy. The effects of non-destructive rock index tests were examined through the use of multiple linear regression (MLR) and adaptive neuro-fuzzy inference system (ANFIS) approaches. Different strategies and scenarios were implemented during modelling of MLR and ANFIS approaches, where the focus was to consider the most important parameters of these techniques. As a result, and according to background and behaviour of the ANFIS (or neuro-fuzzy) model, the predicted values obtained by this intelligent methodology are closer to the actual BTS compared to MLR which works based on linear statistical rules. For instance, in terms of system error and a-20 index, values of (0.84 and 1.20) and (0.96 and 0.80) were obtained for evaluation parts of ANFIS and MLR techniques, which revealed that the ANFIS model outperforms the MLR in forecasting BTS values. In addition, the same results were obtained through ranking systems by the authors. The neuro-fuzzy developed in this study is a strong technique in terms of prediction capacity and it can be used in the other rock-based projects for solving relevant problems.

Multiplicity Eludes Peer Review: The Case of COVID-19 Research.

Multiplicity arises when data analysis involves multiple simultaneous inferences, increasing the chance of spurious findings. It is a widespread problem frequently ignored by researchers. In this paper, we perform an exploratory analysis of the Web of Science database for COVID-19 observational studies. We examined 100 top-cited COVID-19 peer-reviewed articles based on p-values, including up to 7100 simultaneous tests, with 50% including >34 tests, and 20% > 100 tests. We found that the larger the number of tests performed, the larger the number of significant results (r = 0.87, p < 10−6). The number of p-values in the abstracts was not related to the number of p-values in the papers. However, the highly significant results (p < 0.001) in the abstracts were strongly correlated (r = 0.61, p < 10−6) with the number of p < 0.001 significances in the papers. Furthermore, the abstracts included a higher proportion of significant results (0.91 vs. 0.50), and 80% reported only significant results. Only one reviewed paper addressed multiplicity-induced type I error inflation, pointing to potentially spurious results bypassing the peer-review process. We conclude the need to pay special attention to the increased chance of false discoveries in observational studies, including non-replicated striking discoveries with a potentially large social impact. We propose some easy-to-implement measures to assess and limit the effects of multiplicity.

A modified version of MATLAB application window for predicting the weld bead profile and stress–strain plot of AA5052 CMT weldment using ER4043

The present study deals with the extended version of our previous research work. In this article, for predicting the entire weld bead geometry and engineering stress–strain curve of the cold metal transfer (CMT) weldment, a MATLAB based application window (second version) is developed with certain modifications. In the first version, for predicting the entire weld bead geometry, apart from weld bead characteristics, x and y coordinates (24 from each) of the extracted points are considered. Finally, in the first version, 53 output values (five for weld bead characteristics and 48 for x and y coordinates) are predicted using both multiple regression analysis (MRA) and adaptive neuro fuzzy inference system (ANFIS) technique to get an idea related to the complete weld bead geometry without performing the actual welding process. The obtained weld bead shapes using both the techniques are compared with the experimentally obtained bead shapes. Based on the results obtained from the first version and the knowledge acquired from literature, the complete shape of weld bead obtained using ANFIS is in good agreement with the experimentally obtained weld bead shape. This motivated us to adopt a hybrid technique known as ANFIS (combined artificial neural network and fuzzy features) alone in this paper for predicting the weld bead shape and engineering stress–strain curve of the welded joint. In the present study, an attempt is made to evaluate the accuracy of the prediction when the number of trials is reduced to half and increasing the number of data points from the macrograph to twice. Complete weld bead geometry and the engineering stress–strain curves were predicted against the input welding parameters (welding current and welding speed), fed by the user in the MATLAB application window. Finally, the entire weld bead geometries were predicted by both the first and the second version are compared and validated with the experimentally obtained weld bead shapes. The similar procedure was followed for predicting the engineering stress–strain curve to compare with experimental outcomes.

Extending DFA-based multiple linear regression inference: Application to acoustic impedance models

In this paper we propose three distinct points for improving scale sensitive DFA based multiple linear regression inference: the scale dependent standardized Regression Coefficients βˆstdDFAs as a measure of dependent variables relative importance, scale dependent effect size f2(s), and Intersection-Union hypothesis test, which can handle the composite hypothesis of no cross-correlation, and of having no advantage over the standard OLS linear regression model. We applied this framework on a model for acoustic impedance of well log data, as a function of neutron effective porosity, shale volume, and resistivity. We find that the neutron effective porosity is more important for modeling acoustic impedance than resistivity and, that both are more important than the shale volume, for all scales. Results from Intersection-Union test suggest the rejection of compound null hypothesis for neutron effective porosity and resistivity for scales between 40 and 500 ft, suggesting a robust model for acoustic impedance based on neutron effective porosity and resistivity on these scales. This approach represents a novel framework for scale sensitive regression models, and we believe that it can be useful for time series studies in many diverse areas.

Evolutionary relationships among the snakelike pygopodid lizards: a review of phylogenetic studies of an enigmatic Australian adaptive radiation.

Here, I review phylogenetic studies of the lizard family Pygopodidae, a group of 47 extant species that diversified in Australia and New Guinea. The goal of this study was to examine published phylogenetic and phylogenomic hypotheses on pygopodids to identify the strengths and weaknesses in our understanding of their phylogeny. Many parts of the pygopodid family tree are well established by multiple independent tree inferences including: (1) all multispecies genera (i.e., Aprasia, Delma, Lialis, Pletholax, and Pygopus) are monophyletic groups; (2) the root of the pygopodid tree is located along the branch leading to the Delma clade, thus showing that Delma is the sister group to all other pygopodid genera; (3) the Aprasia repens group, Delma tincta group, and several other groups of closely related species are demonstrated to be monophyletic entities; and (4) the monotypic Paradelma orientalis is the sister lineage to the Pygopus clade. Based on accumulated phylogenetic evidence, two taxonomic recommendations are given: Paradelma merits generic status rather than being subsumed into Pygopus as some earlier studies had suggested, and the monotypic Aclys concinna should be recognized as a member of Delma (following current practice) until future studies clarify its placement inside or outside the Delma clade. One chronic problem with phylogenetic studies of pygopodids, which has limited the explanatory power of many tree hypotheses, concerns the undersampling of known species. Although the continual addition of newly described species, especially over the past two decades, has been a major reason for these taxon sampling gaps, deficits in species sampling for ingroups and/or outgroups in several studies of pygopodid species complexes has confounded the testing of some ingroup monophyly hypotheses. Ancient hybridization between non-sister lineages may also be confounding attempts to recover the relationships among pygopodids using molecular data. Indeed, such a phenomenon can explain at least five cases of mito-nuclear discordance and conflicts among trees based on nuclear DNA datasets. Another problem has been the lack of consensus on the relationships among most pygopodid genera, an issue that may stem from rapid diversification of these lineages early in the group’s history. Despite current weaknesses in our understanding of pygopodid phylogeny, enough evidence exists to clarify many major and minor structural parts of their family tree. Accordingly, a composite tree for the Pygopodidae was able to be synthesized. This novel tree hypothesis contains all recognized pygopodid species and reveals that about half of the clades are corroborated by multiple independent tree hypotheses, while the remaining clades have less empirical support.

BOOTSTRAP INFERENCE FOR MULTIPLE CHANGE-POINTS IN TIME SERIES

In this paper, we propose two bootstrap procedures, namely parametric and block bootstrap, to approximate the finite sample distribution of change-point estimators for piecewise stationary time series. The bootstrap procedures are then used to develop a generalized likelihood ratio scan method (GLRSM) for multiple change-point inference in piecewise stationary time series, which estimates the number and locations of change-points and provides a confidence interval for each change-point. The computational complexity of using GLRSM for multiple change-point detection is as low as $O(n(\log n)^{3})$ for a series of length n. Extensive simulation studies are provided to demonstrate the effectiveness of the proposed methodology under different scenarios. Applications to financial time series are also illustrated.

Multi-objective optimization technique for trajectory planning of multi-humanoid robots in cluttered terrain

Humanoid robots hold a decent advantage over wheeled robots because of their ability to mimic human exile. The presented paper proposes a novel strategy for trajectory planning in a cluttered terrain using the hybridized controller modeled on the basis of modified MANFIS (multiple adaptive neuro-fuzzy inference system) and MOSFO (multi-objective sunflower optimization) techniques. The controller works in a two-step mechanism. The input parameters, i.e., obstacle distances and target direction, are first fed to the MANFIS controller, which generates a steering angle in both directions of an obstacle to dodge it. The intermediate steering angles are obtained based on the training model. The final steering angle to avoid obstacles is selected based on the direction of the target and additional obstacles in the path. It is further works as input for the MOSFO technique, which provides the ultimate steering angle. Using the proposed technique, various simulations are carried out in the WEBOT simulator, which shows a deviation under 5% when the results are validated in real-time experiments, revealing the technique to be robust. To resolve the complication of providing preference to the robot during deadlock condition in multi-humanoids system, the dining philosopher controller is implemented. The efficiency of the proposed technique is examined through the comparisons with the default controller of NAO based on toques produces at various joints that present an average improvement of 6.12%, 7.05% and 15.04% in ankle, knee and hip, respectively. It is further compared against the existed navigational strategy in multiple robot systems that also displays an acceptable improvement in travel length. In comparison in reference to the existing controller, the proposed technique emerges to be a clear winner by portraying its superiority.

Aurora: A Generalized Retrieval Framework for Exoplanetary Transmission Spectra

Abstract Atmospheric retrievals of exoplanetary transmission spectra provide important constraints on various properties, such as chemical abundances, cloud/haze properties, and characteristic temperatures, at the day–night atmospheric terminator. To date, most spectra have been observed for giant exoplanets due to which retrievals typically assume hydrogen-rich atmospheres. However, recent observations of mini Neptunes/super-Earths, and the promise of upcoming facilities including the James Webb Space Telescope (JWST), call for a new generation of retrievals that can address a wide range of atmospheric compositions and related complexities. Here we report Aurora, a next-generation atmospheric retrieval framework that builds upon state-of-the-art architectures and incorporates the following key advancements: (a) a generalized compositional retrieval allowing for H-rich and H-poor atmospheres, (b) a generalized prescription for inhomogeneous clouds/hazes, (c) multiple Bayesian inference algorithms for high-dimensional retrievals, (d) modular considerations for refraction, forward scattering, and Mie scattering, and (e) noise modeling functionalities. We demonstrate Aurora on current and/or synthetic observations of the hot Jupiter HD 209458 b, mini Neptune K2-18b, and rocky exoplanet TRAPPIST-1 d. Using current HD 209458 b spectra, we demonstrate the robustness of our framework and cloud/haze prescription against assumptions of H-rich/H-poor atmospheres, improving on previous treatments. Using real and synthetic spectra of K2-18b, we demonstrate an agnostic approach to confidently constrain its bulk atmospheric composition and obtain precise abundance estimates. For TRAPPIST-1 d, 10 JWST-NIRSpec transits can enable identification of the main atmospheric component for cloud-free, CO2-rich, and N2-rich atmospheres and abundance constraints on trace gases, including initial indications of O3 if present at enhanced levels (∼10×–100× Earth levels).

LARGE-SCALE MULTIPLE INFERENCE OF COLLECTIVE DEPENDENCE WITH APPLICATIONS TO PROTEIN FUNCTION.

Measuring the dependence of k ≥ 3 random variables and drawing inference from such higher-order dependences are scientifically important yet challenging. Motivated here by protein coevolution with multivariate categorical features, we consider an information theoretic measure of higher-order dependence. The proposed collective dependence is a symmetrization of differential interaction information which generalizes the mutual information of a pair of random variables. We show that the collective dependence can be easily estimated and facilitates a test on the dependence of k ≥ 3 random variables. Upon carefully exploring the null space of collective dependence, we devise a Classification-Assisted Large scaLe inference procedure to DEtect significant k-COllective DEpendence among d ≥ k random variables, with the false discovery rate controlled. Finite sample performance of our method is examined via simulations. We apply this method to the multiple protein sequence alignment data to study the residue or position coevolution for two protein families, the elongation factor P family and the zinc knuckle family. We identify novel functional triplets of amino acid residues, whose contributions to the protein function are further investigated. These confirm that the collective dependence does yield additional information important for understanding the protein coevolution compared to the pairwise measures.

Simultaneous confidence tubes for comparing several multivariate linear regression models.

Much of the research on multiple comparison and simultaneous inference in the past 60 years or so has been for the comparisons of several population means. Spurrier seems to have been the first to investigate multiple comparisons of several simple linear regression lines using simultaneous confidence bands. In this paper, we extend the work of Liu etal. for finite comparisons of several univariate linear regression models using simultaneous confidence bands to finite comparisons of several multivariate linear regression models using simultaneous confidence tubes. We show how simultaneous confidence tubes can be constructed to allow more informative inferences for the comparison of several multivariate linear regression models than the current approach of hypotheses testing. The methods are illustrated withexamples.

A control structure for ambidextrous robot arm based on Multiple Adaptive Neuro‐Fuzzy Inference System

This paper presents the novel design of an ambidextrous robot arm that offers double range of motion as compared to dexterous arms. The arm is unique in terms of design (ambidextrous feature), actuation (use of two different actuators simultaneously: Pneumatic Artificial Muscle (PAM) and Electric Motors)) and control (combined use of Proportional Integral Derivative (PID) with Neural Network (NN) and Multiple Adaptive Neuro-fuzzy Inference System (MANFIS) controller with selector block). In terms of ambidextrous robot arm control, a solution based on forward kinematic and inverse kinematic approach is presented, and results are verified using the derived equation in MATLAB. Since solving inverse kinematics analytically is difficult, Adaptive Neuro Fuzzy Inference system (ANFIS) is developed using ANFIS MATLAB toolbox. When generic ANFIS failed to produce satisfactory results due to ambidextrous feature of the arm, MANFIS with a selector block is proposed. The efficiency of the ambidextrous arm has been tested by comparing its performance with a conventional robot arm. The results obtained from experiments proved the efficiency of the ambidextrous arm when compared with conventional arm in terms of power consumption and stability.

Dynamic resource allocation for jointing vehicle-edge deep neural network inference

The emergence of mobile edge computing provides an efficient and stable computing platform for intelligent applications of autonomous vehicles, and deep neural network (DNN) based tasks collaborative inference through joint device-edge is considered an effective way to reduce latency. However, the computing resources allocated to the vehicle are dynamic as the number of requesters changes due to the limitation of edge server resources, which causes the best partition point of the DNN is not fixed. In this paper, we consider a dynamic resource allocation scheme to select the best partition point of DNN inference tasks by vehicle-edge collaborative computing. Specifically, the latency constrained DNN tasks of vehicles are partially offloaded to edge at the granularity of DNN layers. Considering the heterogeneity of vehicular computing capabilities and multiple DNN inference tasks, we formulate an optimization problem for dynamic resource allocation and automatically select the best partition point to minimize the overall latency of all vehicles, which is NP-hard. Then we design a chemical reaction optimization based algorithm for low complexity to solve the problem. The results of extensive evaluations illustrate that our proposed scheme is superior to other baseline schemes in terms of overall latency, and with lower failure rate.

“What an idiot!” – How the appraisal of the writer of an uncivil comment impacts discussion behavior

Despite incivility in online discussions being linked to various negative effects, less is known about the mechanisms of how incivility works. So far, explanations by social perception have been neglected. Therefore, drawing on the multiple inference model, this study employs an attribution theoretical approach to examine whether the motives and traits that people attribute to senders of uncivil or opposing comments affect their intentions to join a discussion. Employing a 2 (incivility vs. no incivility) × 2 (like-minded vs. opposing stance) between-subjects design, data from an online experiment ( N = 452) were analyzed applying a path model (SPSS AMOS). Results revealed that participants attributed more aggressive and less nonaggressive motives to senders of uncivil messages. The attribution of aggressive motives consequently increased hostile inferences about the target. A similar pattern occurred when individuals were exposed to an opposing stance. In result, hostile inferences about the sender’s traits decreased participants’ willingness to discuss.

Population Genomics of the "Arcanum" Species Group in Wild Tomatoes: Evidence for Separate Origins of Two Self-Compatible Lineages.

Given their diverse mating systems and recent divergence, wild tomatoes (Solanum section Lycopersicon) have become an attractive model system to study ecological divergence, the build-up of reproductive barriers, and the causes and consequences of the breakdown of self-incompatibility. Here we report on a lesser-studied group of species known as the “Arcanum” group, comprising the nominal species Solanum arcanum, Solanum chmielewskii, and Solanum neorickii. The latter two taxa are self-compatible but are thought to self-fertilize at different rates, given their distinct manifestations of the morphological “selfing syndrome.” Based on experimental crossings and transcriptome sequencing of a total of 39 different genotypes from as many accessions representing each species’ geographic range, we provide compelling evidence for deep genealogical divisions within S. arcanum; only the self-incompatible lineage known as “var. marañón” has close genealogical ties to the two self-compatible species. Moreover, there is evidence under multiple inference schemes for different geographic subsets of S. arcanum var. marañón being closest to S. chmielewskii and S. neorickii, respectively. To broadly characterize the population-genomic consequences of these recent mating-system transitions and their associated speciation events, we fit demographic models indicating strong reductions in effective population size, congruent with reduced nucleotide and S-locus diversity in the two independently derived self-compatible species.