30472 research outputs found
Sort by
Mapping the link between socio-economic factors, autistic traits and mental health across different settings
Get PDFAutistic individuals experience higher rates of externalising and internalising symptoms that may vary with environmental factors. However, there is limited research on variation across settings that may highlight common factors with globally generalisable effects. Data were taken from two cohorts: a multinational European sample (n = 764; 453 autistic; 311 non-autistic; 6–30 years), and a South African sample (n = 100 non-autistic; 3–11 years). An exploratory factor analysis aggregated clinical (Verbal Comprehension and Perceptual Index), adaptive traits (Vineland Adaptive Behaviour Scale) and socio-economic variables (parental employment and education, home and family characteristics) in each cohort separately. With regression, we investigated the effect of these factors and autistic traits on internalising and externalising scores (measured with the Strengths and Difficulties Questionnaire). Cohorts showed similar four-factor structures (Person Characteristics, Family System, Parental and Material Resources). The ‘Family System’ factor captured family size and maternal factors and was associated with lower internalising and externalising symptoms in both cohorts. In the European cohort, high autistic traits reduced this effect; the opposite was found in the South Africa cohort. Our exploratory findings from two separate analyses represent consistent evidence that Family System is associated with internalising and externalising symptoms, with a context-specific impact in persons with high autism traits
Precise quantitative analysis of binarized neural networks: a BDD-based approach
Get PDFAs a new programming paradigm, neural-network-based machine learning has expanded its application to many real-world problems. Due to the black-box nature of neural networks, verifying and explaining their behavior are becoming increasingly important, especially when they are deployed in safety-critical applications. Existing verification work mostly focuses on qualitative verification, which asks whether there exists an input (in a specified region) for a neural network such that a property (e.g., local robustness) is violated. However, in many practical applications, such an (adversarial) input almost surely exists, which makes a qualitative answer less meaningful. In this work, we study a more interesting yet more challenging problem, i.e., quantitative verification of neural networks, which asks how often a property is satisfied or violated. We target binarized neural networks (BNNs), the 1-bit quantization of general neural networks. BNNs have attracted increasing attention in deep learning recently, as they can drastically reduce memory storage and execution time with bit-wise operations, which is crucial in recourse-constrained scenarios, e.g., embedded devices for Internet of Things. Toward quantitative verification of BNNs, we propose a novel algorithmic approach for encoding BNNs as Binary Decision Diagrams (BDDs), a widely studied model in formal verification and knowledge representation. By exploiting the internal structure of the BNNs, our encoding translates the input-output relation of blocks in BNNs to cardinality constraints, which are then encoded by BDDs. Based on the new BDD encoding, we develop a quantitative verification framework for BNNs where precise and comprehensive analysis of BNNs can be performed. To improve the scalability of BDD encoding, we also investigate parallelization strategies at various levels. We demonstrate applications of our framework by providing quantitative robustness verification and interpretability for BNNs. An extensive experimental evaluation confirms the effectiveness and efficiency of our approach
PG-Schema: schemas for property graphs
Get PDFProperty graphs have reached a high level of maturity, witnessed by multiple robust graph database systems as well as the ongoing ISO standardization effort aiming at creating a new standard Graph Query Language (GQL). Yet, despite documented demand, schema support is limited both in existing systems and in the first version of the GQL Standard. It is anticipated that the second version of the GQL Standard will include a rich DDL. Aiming to inspire the development of GQL and enhance the capabilities of graph database systems, we propose PG-Schema, a simple yet powerful formalism for specifying property graph schemas. It features PG-Schema with flexible type definitions supporting multi-inheritance, as well as expressive constraints based on the recently proposed PG-Keys formalism. We provide the formal syntax and semantics of PG-Schema, which meet principled design requirements grounded in contemporary property graph management scenarios, and offer a detailed comparison of its features with those of existing schema languages and graph database systems
Threshold Queries
Get PDFThreshold queries are an important class of queries that only require computing or counting answers up to a specified threshold value. To the best of our knowledge, threshold queries have been largely disregarded in the research literature, which is surprising considering how common they are in practice. We explore how such queries appear in practice and present a method that can be used to significantly improve the asymptotic bounds of their state-of-the-art evaluation algorithms. Our experimental evaluation of these methods shows order-of-magnitude performance improvements
Agentic interplay between hybridity and liminality in contemporary boundary work
Get PDFThere is no doubt that digital technologies have an impact on boundary work, i.e., the practices individuals develop to work up boundaries between ‘work’ and ‘life’. However, related research is still dominated by the integration–segmentation framework which may restrict our understanding of contemporary practice. To address this limitation, we use the concepts of ‘hybridity’ (fusion) and liminality (in-betweenness) in combination to explore how their interplay may promote more nuanced accounts of how individuals work up their boundaries in contemporary everyday life. Our analysis of video diaries and follow-up interviews undertaken by 30 UK–based workers reveals an ‘agentic interplay’ of hybridity and liminality that is understood as an ongoing dynamic practice, sometimes enabling and sometimes undermining desired boundary work outcomes. We make three contributions, collectively offering a critical advancement in the work-life debate and the hybridity and liminality literatures: (a) further developing an understanding of boundary work as situated, creative and dynamic; (b) unpacking the agentic potential of the ‘black box’ of the hybridity–liminality interplay; and (c) expanding current understandings and applications of the concepts of hybridity and liminality
Methodological implications of participant and researcher multilingualism: making language dynamics visible
Get PDFLanguage can impact the research process in complex ways. This special issue (SI) brings together seven contributions which discuss the methodological implications of researching in a multilingual world, where researchers and/or research participants are likely to know more than one language. The papers examine the relationship between researchers’ language ideologies and actual practices with multilingual participants, teams or projects, from a range of theoretical and methodological perspectives. The SI also considers positionality, including how researchers feel–and are perceived–when working in one or more first languages (L1), a later learned language (LX), or a language outside of their linguistic repertoire (L0). It provides practical examples of the stages of researching multilingually, focusing on key decisions that researchers make over the course of their projects, which are seldom made visible in research reports. We argue that linguistic reflexivity is an essential practice, through which researchers make informed language-related choices and continually reflect on the role of language(s) throughout their research projects. These illustrative accounts, from various geographic contexts, offer lessons from experience–distilled as questions and principles–to guide researchers in applied linguistics and beyond as they embark on the multifaceted journey of researching multilingually
Enhancing post‐COVID-19 work resilience in hospitality: a micro-level crisis management framework
Get PDFCOVID-19 has led to immense impacts on the hospitality sector, putting millions of jobs at risk. The uncertainty has resulted in significant psychological effects on service staff. Developing on micro-level interpersonal interactions between line managers and service staff, this article aims to propose a bottom-up approach to alleviate service staff’s anxiety and enhance their work resilience during and after the pandemic. The proposed conceptual framework synthesised the personal uncertainty constructs, self-determination theory (SDT) and leader-member exchange (LMX) to address the psychological issues that emerged from a prolonged ambiguous and unsettled period. The framework proposes that interpersonal exchanges at the micro-level can strengthen service staff’s intrinsic motivation and resilience. This article offers new insights into crisis management research and practice in the hospitality sector by developing a people-centred approach. At the micro-level, this framework offers a holistic strategy, including a flat team structure, uncertainty-embracing organisational environment, transparent decision-making process and sympathetic behaviours from leaders to facilitate service staff to develop greater coping mechanisms for the unknown future. Furthermore, this framework provides organisations preliminary guidelines to establish a longitudinal preventive strategy to strengthen employees’ work resilience
Singapore’s death penalty for drug trafficking: what the research says and doesn’t
No full textMAI SATO reviews studies conducted or commissioned by Singapore’s Ministry of Home Affairs that claim public backing for and the effectiveness of the death penalty in managing drug trafficking. Sato finds that these studies provide far weaker evidence for using the death penalty for drug trafficking than their authors and officials citing them claim
Data sources:
Access Repository Dashboard
Do you manage Birkbeck Institutional Research Online? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!