Search

Background: Operating schools safely under pandemic conditions is a widespread policy goal. We analyse the effectiveness of classroom cohorting, i.e., the decomposition of classrooms into smaller isolated units, in inhibiting the spread of SARS-CoV-2 in European secondary schools and compare different cohorting strategies. Methods: Using real-world network data on 12,291 adolescents collected in classrooms in England, Germany, the Netherlands, and Sweden in 2010/2011, we apply agent-based simulations to compare the effect of forming cohorts randomly to network-based cohorting. Network-based cohorting attempts to allocate out-of-school contacts to the same cohort to prevent cross-cohort infection more effectively. We consider explicitly minimizing out-of-school cross-cohort contacts, approximating this information-heavy optimization strategy by chained nominations of contacts, and dividing classrooms by gender. We also compare the effect of instructing cohorts in-person every second week to daily but separate in-person instruction of both cohorts. Findings: We find that cohorting reduces the spread of SARS-CoV-2 in classrooms. Relative to random cohorting, network-based strategies further reduce infections and quarantines when transmission dynamics are strong. In particular, network-based cohorting inhibits superspreading in classrooms. Cohorting that explicitly minimizes cross-cohort contacts is most effective, but approximation based on chained nominations and classroom division by gender also outperform random cohorting. Every-second-week instruction in-person contains outbreaks more effectively than daily in-person instruction of both cohorts. Interpretation: Cohorting of school classes can curb SARS-CoV-2 outbreaks in the school context. Factoring in out-of-school contacts can achieve a more effective separation of cohorts. Network-based cohorting reduces the risk of outbreaks in schools and can prevent superspreading events. Funding: None.

Clinical studies with time to event endpoints typically report the median follow-up (i.e., censoring) time for the subjects in the trial, alongside the median time to event. The reason for this is to provide information about the opportunity for subjects in the study to experience the event of interest (Betensky, 2015 [1]). The median follow-up time is often calculated from the Kaplan-Meier estimate for time to censoring. In most clinical studies, the censoring time is a composite measure, defined as the minimum of time to drop-out from the study and time to administrative end of study. The time to drop-out component may or may not be observed; it is observed only if drop-out occurs before the event and the end of the study. However, the time to end of study is observed for each subject, as it is the time from entry to the study to the calendar date that is administratively set as the end of the study. It is known even for subjects who have the event prior to the end of the study. This decomposition of the censoring time into a time that is itself potentially censored and a time that is fully observed raises the interesting question of whether estimation of the censoring distribution could be improved through a decoupling of these times. We demonstrate in simulations that consideration of censoring in this way yields reduced variability under some circumstances and should be used in practice. We illustrate these concepts through application to a meningioma study.

This paper presents a basic introduction to model building and identification. Both fundamentals of mathematical description of dynamic systems and the principles of theoretical modelling are emphasized. When familiar with these ideas, the newcomer will understand identification better. Later on in this paper, only a classification of identification methods is given. The other tutorials in this volume will then deal with the methods themselves.

Within the past few decades, the application of light scattering techniques to a broad range of scientific disciplines has increased significantly, especially in the field of analytical chemistry. The resulting interest in and use of light scattering methods suggests the need for an easily understood introduction and review of material for those new to the method as well as for current users in need of a refresher. In many respects, the theory and its applications may appear so overwhelming for many studying the field for the first time, that they rarely can spend the time just needed to understand the basic measurements and their interpretations. A variety of applications in analytical chemistry especially have resulted in a greater understanding of many of the macromolecular processes themselves from molar mass distributions, to the macromolecular interactions responsible for aggregation processes, to determinations of structure and function. The use of such analytical processes to obtain a better understanding of nanoparticle structure and function has become almost universal., Graphical abstract Image 1, Highlights • Deriving the distribution of fractionated macromolecules by light scattering. • Deriving the distribution of fractionated nanoparticles by light scattering. • Deriving the masses of fractionated macromolecules from their scattered light. • Deriving the size of nanoparticles from the angular variation of scattered light. • Deriving bacterial size and structure by light scattering.

Background Hereditary hearing loss (HHL) is the most common sensory deficit, which highly afflicts humans. With gene sequencing technology development, more variants will be identified and support genetic diagnoses, which is difficult for human experts to diagnose. This study aims to develop a machine learning-based genetic diagnosis model of HHL-related variants of GJB2, SLC26A4 and MT-RNR1. Methods This case-control study included 1898 subjects, among which 1354 were HHL patients and 544 were carriers. Risk assessment models were established based on variants at 144 sites in three genes related to HHL by building six machine learning (ML) models. We compared the ML models with the genetic risk score (GRS) and expert interpretation (EI) to verify the clinical performance. Findings Among the six ML models, the support vector machine (SVM) showed the best performance. For the prediction of HHL-related gene sites in subjects with variants, the area under the receiver operating characteristic (AUC) of the SVM model was 0.803 (0.680–0.814) in the 10-fold stratified cross-validation and 0.751 (0.635–0.779) in external validation. The predicted results were better than both EI and GRS. Furthermore, 11 sites were identified as the smallest feature set that can be accurately predicted. Interpretation The developed SVM model has great potential to be an efficient and effective tool for HHL prediction when high throughput sequencing data are available. Funding This study was supported by the National Key Research and Development Program (2017YFC1001800).

This paper presents a novel evaluation approach for optical coherence tomography (OCT) image analysis of pharmaceutical solid dosage forms based on deep convolutional neural networks (CNNs). As a proof of concept, CNNs were applied to image data from both, in- and at-line OCT implementations, monitoring film-coated tablets as well as single- and multi-layered pellets. CNN results were compared against results from established algorithms based on ellipse-fitting, as well as to human-annotated ground truth data. Performance benchmarks used include, efficiency (computation speed), sensitivity (number of detections from a defined test set) and accuracy (deviation from the reference method). The results were validated by comparing the output of several algorithms to data manually annotated by human experts and microscopy images of cross-sectional cuts of the same dosage forms as a reference method. In order to guarantee comparability for all results, the algorithms were executed on the same hardware. Since modern OCT systems must operate under real-time conditions in order to be implemented in-line into manufacturing lines, the necessary steps are discussed on how to achieve this goal without sacrificing the algorithmic performance and how to tailor a deep CNN to cope with the high amount of image noise and alterations in object appearance. The developed deep learning approach outperforms static algorithms currently available in pharma applications with respect to performance benchmarks, and represents the next level in real time evaluation of challenging industrial OCT image data., Graphical abstract Unlabelled Image

After the presentation of various identification and parameter estimation methods in the previous papers, some selected practical aspects of process identification are discussed. This includes, for a given identification method, the steps from the design of experiments to the verification of the final model. Therefore a general procedure of process identification, the selection of input signals, the selection of the sampling time, off-line and on-line identification, comparison of parameter estimation methods, model order testing and model verification is presented. A short discussion on program packages for process identification follows.

This paper surveys the evolution of “high-lever” programming languages during the first decade of computer programming activity. We discuss the contributions of Zuse in 1945 (the “Plankalkul”), Goldstine and von Neumann in 1946 (“Flow Diagrams”), Curry in 1948 (“Composition”), Mauchly et al. in 1949 (“Short Code”), Burks in 1950 (“Intermediate PL”), Rutishauser in 1951 (“Klammerausdrucke”), Bohm in 1951 (“Formules”), Glennie in 1952 (“AUTOCODE”), Hopper et al. in 1953 (“A-2”), Laning and Zierler in 1953 (“Algebraic Interpreter”), Backus et al. in 1954–1957 (“FORTRAN”), Brooker in 1954 (“Mark I AUTOCODE”), Kamynin and Liubimskii in 1954 (“ΠΠ-2”), Ershov in 1955 (“ΠΠ”), Grems and Porter in 1955 (“BACAIC”), Elsworth et al. in 1955 (“Kompiler 2”), Blum in 1956 (“ADES”), Perlis et al. in 1956 (“IT”), Katz et al. in 1956–1958 (“MATH-MATIC”), Bauer and Samelson in 1956–1958 (U.S. Patent 3,047,228). The principal features of each contribution are illustrated and discussed. For purposes of comparison, a particular fixed algorithm has been encoded (as far as possible) in each of the languages. This research is based primarily on unpublished source materials, and the authors hope that they have been able to compile a fairly complete picture of the early developments in this area.

The bond portfolio problem is viewed as a multistage decision problem in which buy, sell, and hold decisions are made at successive (discrete) points in time. Normative models of this decision problem tend to become very large, particularly when its dynamic structure and the uncertainty of future interest rates and cash flows are incorporated in the model. In this paper we present a multiple period bond portfolio model and suggest a new approach for efficiently solving problems which are large enough to make use of as much information as portfolio managers can reasonably provide. The procedure utilizes the decomposition algorithm of mathematical programming and an efficient technique developed for solving subproblems of the overall portfolio model. The key to the procedure is the definition of subproblems which are easily solved via a simple recursive relationship.

Despite appealing characteristics for the clinical trials setting, Bayesian inference methods remain scarcely used, especially in randomized controlled clinical trials (RCT). This is particularly true when dealing with a survival endpoint, likely due to the additional complexities to model specifications. We propose to use Bayesian inference to estimate the treatment effect in this setting, using a proportional hazards (PH) model for right-censored data. Implementation of such an estimation process is illustrated on two working examples from cancer RCTs, the ALLOZITHRO and the CLL7-SA trials, both originally analyzed using a frequentist approach. In these two different settings, we show that Bayesian sequential analyses can provide early insight on treatment effect in RCTs. Relying on posterior distributions and predictive posterior probabilities, we find that Bayesian sequential analyses of the ALLOZITHRO trial, which was terminated early due to an unanticipated deleterious effect of the intervention on survival, allow quantifying early that the treatment effect was opposite to what was expected. Then, incorporating historical data in the sequential analyses of the CLL7-SA trial would have allowed the treatment effect to be closer to the protocol hypothesis. These post-hoc results give grounds to advocate for a wider use of Bayesian approaches in RCTs, including those with right-censored endpoints, as informative decision tools.

Publisher Summary The choice of an individual decision maker among alternative risky ventures may be regarded as a two-step procedure. The decision maker chooses an efficient set among all available portfolios, independently of his tastes or preferences. Then, the decision maker applies individual preferences to this set to choose the desired portfolio. The subject of this chapter is the analysis of the first step. It deals with optimal selection rules that minimize the efficient set by discarding any portfolio that is inefficient in the sense that it is inferior to a member of the efficient set, from point of view of each and every individual, when all individuals' utility functions are assumed to be of a given general class of admissible functions. The analysis presented in the chapter is carried out in terms of a single dimension such as money, both for the utility functions and for the probability distributions. However, the results may easily be extended, with minor changes in the theorems and the proofs, to the multivariate case. The chapter explains a necessary and sufficient condition for efficiency, when no further restrictions are imposed on the utility functions. It presents proofs of the optimal efficiency criterion in the presence of general risk aversion, that is, for concave utility functions.

Publisher Summary This chapter supplements Professor Hare's presentation on experimentation on children by widening the context. The chapter presents the distinction between two main areas of medical ethics: clinical medical ethics (the ethics of clinical medical practice), and the ethics of biomedical research, a distinction that is presupposed in Professor Hare's paper. Biomedical research ethics presupposes the ethics of clinical medical practice, which has its foundations in our common and shared morality. One conclusion to be drawn from this is that although, in some sense, medical ethics is specific and different from “ordinary ethics”— the commonly shared morality prevailing in a given society—neither branch of medical ethics is autonomous or independent of general morality. Professor Hare's paper is held in a general and abstract vein in the sense that it contains few references to case material. According to Professor Hare, the distinction between deontological and utilitarian moral theory and argumentation is of importance in medical ethics, and certainly when it comes to experimentation on children.

Publisher Summary This chapter discusses two main revolutions in the computer field. The first was the employment of vacuum tubes to make a fast, reliable, powerful, general-purpose computer. This development began with John Atanasoff's slow, special-purpose electronic computer called ENIAC. The second revolution was the stored-program computer. The ENIAC necessarily used electromechanical equipment for input and output, and thereby tested the relative merits of the two technologies for computing. It is claimed that the arithmetic design of ENIAC was influenced mainly by two kinds of calculators: mechanical desk calculators, electrically powered and hand operated; and electromechanical card operated IBM machines. During 1946, the final design and construction stage of the development of the stored-program computer was beginning. It is suggested that von Neumann was the first to see and exploit the fact that when orders or instructions are stored in a high-speed read–write electronic memory, they can be manipulated arithmetically and modified by the machine. His variable address EDVAC code was the basis of the modern computer software revolution.

Publisher Summary This chapter explains techniques for increasing a layperson's view of computers and computer programming. It discusses the usefulness of five recommendations concerning the design of computer literacy curricula: (1) provide the learner with a concrete model of the computer; (2) encourage the learner to actively restate the new technical information in his or her own words; (3) assess the learner's existing intuitions about computer operation and try to build on them, or modify them, as needed; (4) provide the learner with methods for chunking statements into a larger, single, meaningful unit; and (5) provide the learner with methods for analyzing statements into smaller, meaningful parts. Each recommendation is concerned with increasing the meaningfulness of learning new computer information by novices. Information enters the human cognitive system from the outside and must go through the following steps: (1) reception, (2) availability, and (3) activation.

Publisher Summary This chapter discusses time consistency and robustness of equilibria in noncooperative dynamic games. It presents a unified description of time consistency in dynamic decision problems, so as to encompass both stochastic and deterministic models, and a variety of information patterns and solution concepts. For a meaningful study of time consistency, it is essential that the problem formulation be consistent across time. One of the major differences between static (or myopic) policies and dynamic ones is that a dynamic policy uses the available a priori information on the future evolution of the decision process such as the decision (state) dynamics, the time horizon, the possibility of acquiring some dynamic information in the future, and the trade-offs between current and future performances. The issue of time consistency of optimal or equilibrium policies in dynamic single-agent and multiple-agent (game) problems is a topic of major importance in economic decision making. Without a reliable probabilistic assessment of deviations from a nominal model, it would not be possible to accommodate such problems in the general framework.

Publisher Summary This chapter presents classroom organization projects that provide particularly, clear illustration of the conditions and strategies that support mutual adaptation and thus, successful implementation. They are especially relevant to understanding the operational implications of the Change Agent study finding for policy and practice not only because mutual adaptation is intrinsic to change in classroom organization, but also because the question of institutional receptivity does not cloud the view of effective implementation strategies afforded by these projects. Overcoming the challenges and problems inherent to innovations in classroom organization contributes positively and significantly to their effective implementation. The amorphous yet highly complex nature of classroom organization projects tends to require or dictate an adaptive implementation strategy that permits goals and methods to be reassessed, refined, and made explicit during the course of implementation, and that fosters learning-by-doing.

Publisher Summary The quantification requires the selection of one of a set of special morphemes that designates certain semantic features of whatever is being enumerated. Changes in classifiers are interesting because they provide an occasion to investigate an idea of traditional concern in linguistics—the relationship between language learning and language change. One can use shape or process classifiers for most of the things that have their own taxonomy specific classifiers. The unmarked classifiers are used in place of the more specific marked forms and also true in situations where there are intergenerational differences in use of forms. Numeral classifiers differ from phonology as they appear infrequently in spontaneous conversation and classifier forms are not among the earliest part of the language mastered because they depend not only on the mastery of number concepts but also on having a set of lexemes representing concepts to which they can be applied.

The full advantages of the incremental properties of production systems have yet to be exploited on a large scale. A promising vehicle for this is the task of instructing a system to solve problems in a complex domain. For this, it is important to express the instruction in a language similar to natural language and without detailed knowledge of the inner structure of the system. Instruction and close interaction with the system as it behaves are preferred over a longer feedback loop with more independent learning by the system. The domain is initially an abstract job shop. The beginning system has capabilities for solving problems, processing language, building productions, and interacting with the task environment. All parts of the system are subject to instruction. The main problem-solving strategy, which permeates all four system components, is based on means-ends analysis and goal-subgoal search. This is coupled with an explicit representation of control knowledge. The system's behavior so far is restricted to simple environmental manipulations, a number of which must be taught before more complex tasks can be done.

A system for the development of the work process, work activity, and the mental models for work in the paper industry was created in this study. A developmental program for the personnel of a paper mill was planned and carried out to eliminate the problems and to develop the work process. During the program the five crews as well as the technicians and engineers working on one paper machine analyzed their own work process in detail. Several problems and developmental needs were recognized. These dealt with the equipment, flow of information, work activities, work division and professional qualifications of the personnel. More than 60 specified problems or developmental needs that came up were solved. The solutions, prerequisites for their realization, persons responsible for carrying out the improvements, and the schedule for the execution were created and documented. This developmental program rested on the activity and knowledlge of the personnel of the paper mill. The researchers structured the program, created the tasks for the participants, and led the discussions. After the intervention the mental models for the work process have become more complete, the assessments of work characteristics are more positive, and the productivity of the machine has increased by about 30%.

In a snoopy cache multiprocessor system, each processor has a cache in which it stores blocks of data. Each cache is connected to a bus used to communicate with the other caches and with main memory. Each cache monitors the activity on the bus and in its own processor and decides which blocks of data to keep and which to discard. For several of the proposed architectures for snoopy caching systems, we present new on-line algorithms to be used by the caches to decide which blocks to retain and which to drop in order to minimize communication over the bus. We prove that, for any sequence of operations, our algorithms' communication costs are within a constant factor of the minimum required for that sequence; for some of our algorithms we prove that no on-line algorithm has this property with a smaller constant.

Publisher Summary This chapter presents a method for the complete finite axiomatization of every finite logic. It resembles a Gentzen rather than a Hilbert-type axiomatization, which means that the process of proving a formula as a formal theorem is reduced to its decomposition into a set of its sub-formulae. But the method is not a kind of a Gentzenization. The original Gentzen-type proof procedures appear to be formalizations of direct intuitive reasoning. The method consists of an indirect reduction of proving a kind of “negation” of the formula to a kind of a “contradiction” and it appears to be a formalization of indirect rather than direct intuitive reasoning.

We analyze the structural elements which comprise a standard scientific paper. Previous analysis generally has focused on one element of a paper at a time. However, the title, author list, affiliation, abstract, text, tables, graphs, charts, photographs and references all represent possible data resources for investigation. After specifying those elements, we focus successively on the history, normative tradition, and sociological analysis of a selection of those elements.

Collaborative approaches in health, such as One Health (OH), are promising; nevertheless, several authors point at persistent challenges for designing and implementing OH initiatives. Among other challenges, OH practitioners struggle in their efforts to collaborate across disciplines and domains. This paper aims to provide insights into the existing challenges for designing and implementing OH initiatives, their causes and solutions, and points out strategic solutions with the potential to solve practical challenges. A systematic literature search was performed for emerging challenges and proposed solutions in the process of conducting OH initiatives. Next, a thematic and a causal analysis were performed to unravel challenges and their causes. Finally, solutions were discriminated on whether they were only recommended, or implemented as a proof-of-principle. The 56 included papers describe 21 challenges endured by OH initiatives that relate to different themes (policy and funding; education and training; surveillance; multi-actor, multi-domain, and multi-level collaborations; and evidence). These challenges occur in three different phases: the acquisition of sufficient conditions to start an initiative, its execution, and its monitoring and evaluation. The findings indicate that individual challenges share overlapping causes and crosscutting causal relations. Accordingly, solutions for the successful performance of OH initiatives should be implemented to tackle simultaneously different types of challenges occurring in different phases. Still, promoting collaboration between the wide diversity of stakeholders, as a fundamental aspect in the OH approach, is still by far the most challenging factor in performing OH initiatives. Causes for that are the difficulties in promoting meaningful and equal participation from diverse actors. Solutions proposed for this challenge focused on guiding stakeholders to think and collaborate beyond their professional and cultural silos to generate knowledge co-creation and innovative methodologies and frameworks. Finally, the biggest knowledge gap identified, in terms of proposed solutions, was for monitoring and evaluating OH initiatives. This highlights the need for future research on evaluation methods and tools specific for the OH approach, to provide credible evidence on its added value. When considering challenges endured by former OH initiatives and the proposed solutions for these challenges, practitioners should be able to plan and structure such initiatives in a more successful way, through the strategic pre-consideration of solutions or simply by avoiding known barriers. Keywords: One health, Challenges, Design, Implementation, Interdisciplinary collaboration, Transdisciplinary research, strategic solutions

We introduce a Bayesian optimization method for estimating the maximum tolerated dose in this article. A number of parametric model-based methods have been proposed to estimate the maximum tolerated dose; however, parametric model-based methods need an assumption that dose-toxicity relationships follow specific theoretical models. This assumption potentially leads to suboptimal dose selections if the dose-toxicity curve is misspecified. Our proposed method is based on a Bayesian optimization framework for finding a global optimizer of unknown functions that are expensive to evaluate while using very few function evaluations. It models dose-toxicity relationships with a nonparametric model; therefore, a more flexible estimation can be realized compared with existing parametric model-based methods. Also, most existing methods rely on point estimates of dose-toxicity curves in their dose selections. In contrast, our proposed method exploits a probabilistic model for an unknown function to determine the next dose candidate without ignoring the uncertainty of posterior while imposing some dose-escalation limitations. We investigate the operating characteristics of our proposed method by comparing them with those of the Bayesian-based continual reassessment method and two different nonparametric methods. Simulation results suggest that our proposed method works successfully in terms of selections of the maximum tolerated dose correctly and safe dose allocations.

Summary: Background: Detailed evaluation of bile duct (BD) is main focus during endoscopic ultrasound (EUS). The aim of this study was to develop a system for EUS BD scanning augmentation. Methods: The scanning was divided into 4 stations. We developed a station classification model and a BD segmentation model with 10681 images and 2529 images, respectively. 1704 images and 667 images were applied to classification and segmentation internal validation. For classification and segmentation video validation, 264 and 517 videos clips were used. For man-machine contest, an independent data set contained 120 images was applied. 799 images from other two hospitals were used for external validation. A crossover study was conducted to evaluate the system effect on reducing difficulty in ultrasound images interpretation. Findings: For classification, the model achieved an accuracy of 93.3% in image set and 90.1% in video set. For segmentation, the model had a dice of 0.77 in image set, sensitivity of 89.48% and specificity of 82.3% in video set. For external validation, the model achieved 82.6% accuracy in classification. In man-machine contest, the models achieved 88.3% accuracy in classification and 0.72 dice in BD segmentation, which is comparable to that of expert. In the crossover study, trainees’ accuracy improved from 60.8% to 76.3% (P

Since the past decade, the use of nanomaterials, especially nano-molecular imprinted (nano-MIP) systems, production and equipment integration are rapidly improving due to affinity sensors. This chapter focuses on the nanobiosensors usability of imprinted nanopolymers for various purposes. It is important to assess potential applications for several areas such as environmental, food safety, medicine, etc. In addition, latest trend is minimally invasive or non-invasive wearable and implantable biomedical applications for continuous and real-time monitored systems. This chapter presents an overview of nano-MIP sensors and classification of nanopolymers and discusses basic properties and potential applications. © 2021 Elsevier Inc.

11 pages, 7 figures, Marine Ecosystem Models (MEMs) provide a deeper understanding of marine ecosystem dynamics. The United Nations Decade of Ocean Science for Sustainable Development has highlighted the need to deploy these complex mechanistic spatial-temporal models to engage policy makers and society into dialogues towards sustainably managed oceans. From our shared perspective, MEMs remain underutilized because they still lack formal validation, calibration, and uncertainty quantifications that undermines their credibility and uptake in policy arenas. We explore why these shortcomings exist and how to enable the global modelling community to increase MEMs’ usefulness. We identify a clear gap between proposed solutions to assess model skills, uncertainty, and confidence and their actual systematic deployment. We attribute this gap to an underlying factor that the ecosystem modelling literature largely ignores: technical issues. We conclude by proposing a conceptual solution that is cost-effective, scalable and simple, because complex spatial-temporal marine ecosystem modelling is already complicated enough, MC and JS acknowledge funding from the EuroMarine 2018 call for Foresight Workshops and Working Groups proposals. JS, MC and MGP acknowledge the Spanish Ministry of Science and Innovation grant agreement N° PID2020-118097RB-I00 (ProOceans). JS, MC and YJS acknowledge the European Union's Horizon 2020 research and innovation programme under grant agreements N° 869300 (FutureMARES) and N° 817578 (TRIATLAS). ADM acknowledges the European Union's Horizon 2020 research and innovation programme under grant agreements N° 869300 (FutureMARES). MC acknowledges the ‘Severo Ochoa Centre of Excellence’ accreditation (CEX2019-000928-S) to the Institute of Marine Science. GR acknowledges the German Federal Ministry of Education and Research (BMBF, Humboldt Tipping project 01LC1823D). VC acknowledges support through NSERC Discovery Grant RGPIN-2019-04901. VC and GO acknowledge funding through the NSERC Mitacs Accelerate Fellowship (IT09266), IOF Ocean Leaders Fellowship, and DFO. YJS acknowledges funding support from the Biodiversa and Belmont Forum project SOMBEE (BiodivScen ERA-Net COFUND programme, ANR contract n°ANR-18-EBI4-0003-01), and the Pew marine fellows programme

Publisher Summary This chapter describes a few strategies for the first two years of language learning. Studies of children's texts and of such other performances as comprehension and imitation have given some basis for testing plausible generalizations about what is necessary for language processing in its early stages. The prerequisites to language are environmental circumstances, cognitive development, and children's information storage. On the basis of these considerations, the basic acquisition of regularities of order and inflection, the fundamental grammatical features found in the early performances of children, can be examined. There are three categories of prerequisites to language learning, the first consisting of what the environment must provide, the others the knowledge that language refers to, and the processing skills that the child brings to the learning task. At the onset of grammar in children, while surface markers of simple structure may facilitate the discovery of processing heuristics, they are not absolutely necessary. The abstracting process that allows children to discover formal similarities, presumably from distributional features, is at least as early in development as the appearance of the auxiliary system in English.

Publisher Summary This chapter discusses the papers presented at the Symposium on Mathematical Software. The papers are grouped in a certain way, but many touch on a variety of area. The theme of the papers is user modifiability. The first of four papers are on arithmetic, and present a treatise on the monitoring of significance in computation and its relation to number representation. The papers present the point of view of people outside the professional community of mathematical software. They give an account of how the lunar mass concentrations were discovered and emphasize the important role that high-quality mathematical software played. The papers are against the black-box approach to library programs and find most library programs out-of-date, from both the mathematical and the systems point of view. They support the development of modular, well-documented programs that the user can modify to suit his own needs.

This short paper presents an HPC enhanced Agent Based Model (ABM) developed with the aim of quantitatively estimating the strategies for accelerating emergency mass evacuations, like tsunami evacuation. In order to facilitate inclusion of various influencing factors, such as localized congestion, multi-modes, pedestrian vehicle interactions, fallen debris from damaged buildings, visibility, etc., which demand detailed models, the developed system includes a 1mx1m resolution model of environment, and agents capable of perceiving and autonomously interact with this high resolution environment and visible agents. In order to meet the computational demand of large scale simulations with complex agents, a scalable high performance extension was implemented. Short introductions to the agent based model and the HPC extension are presented in this paper. In order to demonstrate the scalability of the system, both in problem size and computational capability, a 588 km2 region with 10 million agents is simulated in K computer. It is demonstrated that the system has high strong scalability up to 2048 computing nodes, which is equivalent to 16,384 CPU cores., Transportation Research Procedia, 34, ISSN:2352-1465, International Symposium of Transport Simulation (ISTS’18) and the International Workshop on Traffic Data Collection and its Standardization (IWTDCS’18) Emerging Transport Technologies for Next Generation Mobility

The application of X-ray microtomography for quantitative structural analysis of pharmaceutical multi-particulate systems was demonstrated for commercial capsules, each containing approximately 300 formulated ibuprofen pellets. The implementation of a marker-supported watershed transformation enabled the reliable segmentation of the pellet population for the 3D analysis of individual pellets. Isolated translation- and rotation-invariant object cross-sections expanded the applicability to additional 2D image analysis techniques. The full structural characterisation gave access to over 200 features quantifying aspects of the pellets' size, shape, porosity, surface and orientation. The extracted features were assessed using a ReliefF feature selection method and a supervised Support Vector Machine learning algorithm to build a model for the detection of broken pellets within each capsule. Data of three features from distinct structure-related categories were used to build classification models with an accuracy of more than 99.55% and a minimum precision of 86.20% validated with a test dataset of 886 pellets. This approach to extract quantitative information on particle quality attributes combined with advanced data analysis strategies has clear potential to directly inform manufacturing processes, accelerating development and optimisation., Graphical abstract Unlabelled Image, Highlights • Coupling micro-XRT analysis with feature selection and machine learning for advanced pharmaceutical product characterisation. • Information on particle 3D-orientation were utilised to extract translation- and rotation-invariant object cross-sections. • Successful extraction of over 200 quantitative pellet descriptors linked to size, shape, porosity, surface and orientation. • Sensitivity analysis and ReliefF feature selection approach to identify predictive features for pellet classification. • Feature-based binary SVM classification model for the detection of broken pellets within the formulated system.

Publisher Summary Ever since the introduction of high-speed computers, econometricians have been able to analyze data with great ease. Although this computational revolution has been beneficial on the whole, it has had an unfortunate side effect. One is less likely to uncover erroneous observations and other anomalies in the data when using a computer than when analyzing data by hand. With large data sets, one has almost no choice but to rely entirely on well-defined programmable diagnostics and estimators. In recent years, statisticians have begun to study the existing statistical procedures to see as to which ones have this limited-sensitivity property and to develop alternative techniques where necessary. The papers by Roy Welsch and Bruce Hill provide examples. The simplest way to achieve protection against aberrant data is to use diagnostics that reveal the influential observations. Those observations which have a large effect on the results should be examined closely, keeping in mind the possibility that they were generated by a process not fully accounted for by the model.

Publisher Summary This chapter presents a comparison of biplot display and pencil-and-paper exploratory data analysis (EDA) methods. The chapter also presents an introduction to the biplot accompanied by an example in which the biplot is used for data summarization and description. On a point of terminology, the prefix “bi–” of biplot indicates that this is a joint display of rows and columns. It does not indicate the two-dimensionality of the biplot. The chapter discusses the uses of the biplot. These are mostly of two kinds: (1) inspection of data and (2) diagnostics. In using the biplot, a few displays suffice to reveal relevant patterns in a pretty striking manner. EDA may show more detail if one inspects fits and residuals carefully at each stage but it requires iterative cycles of modelling, fitting, residuals inspection, re-expression, and decisions. If one fit is inadequate, another is tried until a model is judged adequate. This is a search by trial and error rather than by a systematic method. Irregularities are provided by data, inspection takes time, but inspiration is something that may be difficult to come by. The EDA modeling procedure is in general not systematic. Biplot diagnostics are more systematic and direct. One does not start by guessing a model but rather displays data and inspects it, and the diagnosis is then often immediate.

Publisher Summary This chapter presents the views of L. H. Koopmans, Watts, and Folks regarding teaching service courses. Koopmans' and Watts' views are based on their instructional strategies and statistical merits separately in that order. Folks' paper does not lend itself to such a dichotomy and, thus, presents his view on the advantages, disadvantages, and appraisals of service courses. While Koopmans has worked out the subject matter of his introductory statistics course with great care, he has paid little attention to the process of instruction—the modes and strategies of teaching. Watts, in his approach, has failed to mention the constraints that will affect the final instructional design and the resources available to teach the tasks. Folks, in his view, has focused on impressions and opinions rather than clients' needs and rationale for his layout of the service courses.

R 2 D 2 is an object-oriented database system intended for engineering applications that is based on the nested relational data model. R 2 D 2 provides a two layer architecture for engineering application programming: a public database to globally store the engineering objects and a private database consisting of a local database and an object cache that are local to the application program. This paper is concerned with the transaction control mechanisms needed to control concurrent transactions that access the public database system. Objects are requested by a transaction to be transferred from the public database into the transaction's local database connected with an object cache, which together form the private data repository. Upon release of the objects from the cache they are checked back into the public database. The central implementation idea forms a modified intention locking scheme to provide for a high level of concurrency in accessing objects that are represented in nested relational structure, the underlying data model of R 2 D 2 . This scheme facilitates the (exclusive or shared) locking of subobjects within an abstraction hierarchy while the remainder of the hierarchical object is still accessible by other transactions.

A control system for the energy supply to a paper factory needed re-designing to overcome stability problems and uneconomic plant operation. A modelling and simulation study was undertaken with a view to the proposal of a new control structure. The system was simulated using a microprocessor-based colour display system and the BASIC interpretative language with some enhancements. This led to a control proposal which on the simulation overcame the original problems and enabled the desired plant goals to be achieved. Little direct use was made of modern control theory, as this appeared to be inapplicable to the detailed requirements of the problem. Rather, the general principles of modern control and dynamic systems theory led to a heuristic design based on plant understanding and simplified physical cause and effect relationships.

Informatics is usually considered as the science of computer applications in very different domains. The complexity of hardware and software is such that people working in informatics have to be carefully trained. Now, this training is mainly a mathematical training. But informatics has much closer connections to systems theory and cybernetics than to mathematics. It is shown in this paper that informatics is first of all a system science. This conclusion has some influence on the informatics curricula.

In many robotic applications the need to represent and reason about spatial relationships is of great importance. However, our knowledge of particular spatial relationships is inherently uncertain. The most used method for handling the uncertainty is to "pre-engineer" the problem away, by structuring the working environment and using specially-suited high-precision equipment. In some advanced robotic research domains, however, such as automatic task planning, off-line robot programming, and autonomous vehicle operation, prior structuring will not be possible, because of dynamically changing environments, or because of the demand for greater reasoning flexibility. Spatial reasoning is further complicated because relationships are often not described explicitly, but are given by uncertain relative information. This is particularly true when many different frames of reference are used, producing a network of uncertain relationships. Rather than treat spatial uncertainty as a side issue in geometrical reasoning, we believe it must be an intrinsic part of spatial representations. In this paper, we describe a representation for spatial information, called the stochastic map, and associated procedures for building it, reading information from it, and revising it incrementally as new information is obtained. The map always contains the best estimates of relationships among objects in the map, and their uncertainties. The procedures provide a general solution to the problem of estimating uncertain relative spatial relationships. The estimates are probabilistic in nature, an advance over the previous, very conservative, worst-case approaches to the problem. Finally, the procedures are developed in the context of state-estimation and filtering theory, which provides a solid basis for numerous extensions.

This paper presents some problems in communication networks which can be addressed by combinatorial methods. Of particular interest are discrimination problems in radar networks, partitioning of core storage, and data inconsistency.

Graphical Abstract The prediction model of oral disintegrating tablets formulations with direct compression process by artificial neural network (ANN) and deep neural network (DNN) techniques were established. 145 formulation data were extracted from Web of Science. All datasets were divided into three parts: training set (105 data), validation set (20) and testing set (20) to build prediction model.Image, graphical abstract, Oral disintegrating tablets (ODTs) are a novel dosage form that can be dissolved on the tongue within 3 min or less especially for geriatric and pediatric patients. Current ODT formulation studies usually rely on the personal experience of pharmaceutical experts and trial-and-error in the laboratory, which is inefficient and time-consuming. The aim of current research was to establish the prediction model of ODT formulations with direct compression process by artificial neural network (ANN) and deep neural network (DNN) techniques. 145 formulation data were extracted from Web of Science. All datasets were divided into three parts: training set (105 data), validation set (20) and testing set (20). ANN and DNN were compared for the prediction of the disintegrating time. The accuracy of the ANN model have reached 85.60%, 80.00% and 75.00% on the training set, validation set and testing set respectively, whereas that of the DNN model were 85.60%, 85.00% and 80.00%, respectively. Compared with the ANN, DNN showed the better prediction for ODT formulations. It is the first time that deep neural network with the improved dataset selection algorithm is applied to formulation prediction on small data. The proposed predictive approach could evaluate the critical parameters about quality control of formulation, and guide research and process development. The implementation of this prediction model could effectively reduce drug product development timeline and material usage, and proactively facilitate the development of a robust drug product.

In product development, it is crucial to choose the appropriate drug manufacturing route accurately and timely and to ensure that the technique selected is suitable for achieving the desired product quality. Guided by the QbD principles, the pharmaceutical industry is currently transitioning from batch to continuous manufacturing. In this context, process understanding and prediction are becoming even more important. With regard to hot melt extrusion, the process setup, optimization and scale-up in early stages of product development are particularly challenging due to poor process understanding, complex product-process relationship and a small amount of premix available for extensive experimental studies. Hence, automated, quick and reliable process setup and scale-up requires simulation tools that are accurate enough to capture the process and determine the product-process relationships. To this end, the effect of process settings on the degradation of the active pharmaceutical ingredient (API) in a lab-scale Leistritz ZSE12 extruder was investigated. As part of the presented study, the limitations of traditional process analysis using integral process values were investigated, together with the potential that simulations may have in predicting the process performance and the product quality. The results of our investigation indicate that the average melt temperatures and the exposure times in specific zones along the screw configuration correlate well with the API degradation values and can be used as potent process design criteria to simplify the process development., Graphical abstract Unlabelled Image

This chapter discusses duplex design. The primary objectives of the duplex design is to estimate the general score of individual i in group h , θ hi the subarea scores for group h , φ * hj , and the change of the individual-level item parameters over time. The hierarchical model, which is used, assumes random group effects ( θ h , φ hj ) and random individual effects ( θ hi , φ hij ) distributed about the group effects. To reduce the model with many parameters to a form more suitable for producing the desired estimates, a number of simplifying assumptions have been made. In practice, the estimation procedure is carried out in two stages: the first is to estimate the item parameters ( a, b ), and the second is to estimate θ h and φ hj by their posterior expectations, conditionally on the item parameters assumed to equal those estimated from the first stage.

Publisher Summary Turn taking is used for the ordering of moves in games, for allocating political office, for regulating traffic at intersections, for the servicing of customers at business establishments, and for talking in interviews, meetings, debates, ceremonies, conversations. This chapter discusses the turn-taking system for conversation. On the basis of research using audio recordings of naturally occurring conversations, the chapter highlights the organization of turn taking for conversation and extracts some of the interest that organization has. The turn-taking system for conversation can be described in terms of two components and a set of rules. These two components are turn-constructional component and turn-constructional component. Turn-allocational techniques are distributed into two groups: (1) those in which next turn is allocated by current speaker selecting a next speaker and (2) those in which next turn is allocated by self-selection. The turn-taking rule-set provides for the localization of gap and overlap possibilities at transition-relevance places and their immediate environment, cleansing the rest of a turn's space of systematic bases for their possibility.

The type of cognitive system (CS) studied here has four basic parts: (1) a set of interacting elementary productions, called lassifiers, (2) a performance algorithm that directs the action of the system in the environment, (3) a simple learning algorithm that keeps a record of each classifier's success in bringing about rewards, and (4) a more complex learning algorithm, called the genetic algorithm, that modifies the set of classifiers so that variants of good classifiers persist and new, potentially better ones are created in a provably efficient manner. Two “proof-of-principle” experiments are reported. One experiment shows CS's performance in a maze when it has only the ability to adjust the predictions about ensuing rewards of classifiers (similar to adjusting the “weight” of each classifier) vs. when the power of the genetic algorithm is added. Criterion was achieved an order of magnitude more rapidly when the genetic algorithm was operative. A second experiment examines transfer of learning. Placed in a more difficult maze, CS with experience in the simpler maze reaches criterion an order of magnitude more rapidly than CS without prior experience.

We present a new methodology for the design and development of large-scale simulations. Such simulations have often been criticized for their lack of intelligibility, modifiability, credibility, and performance. They are costly to build, to run, to interpret, and to change. Recent advances in artificial intelligence, most notably in the areas of expert systems and object-oriented languages, suggest new approaches to overcoming these limitations. We have applied this technology to the design and implementation of a new simulation language and environment called ROSS (Rule-Oriented Simulation System). Within our research system, we have implemented two military battle simulations, SWIRL (a strategic simulation) and TWIRL (a tactical simulation), which simulate the interactions and outcomes of combat between two opposing military forces. A graphics facility has been developed to visually display simulation runs dynamically. written in Franzlisp running on a VAX. Our graphics routines are written in C and loaded directly into the Franzlisp environment. Our graphics processor is an AED 512. Also we are experimenting with a network of Xerox 1100s (Lisp Machines) for our work on distributed simulation (discussed elsewhere [4]).