107159 research outputs found
Sort by
Universality for degenerate hypergraphs
Get PDFA graph Γ is said to be universal for a class of graphs H if Γ contains a copy of every H ϵ H as a subgraph. The number of edges required for a host graph Γ to be universal for the class of D-degenerate graphs on n vertices has been shown to be O((log n)2/D(log log n)5n2-1/D). We generalise this result to r-uniform hypergraphs, showing the following. Given D, r ≥ 2 and n sufficiently large, there exists a constant C = C(D, r) such that there exists a graph with at most Cnr-1/D(log n)2/D(log log n)2r+1 edges, which is universal for the class of D-degenerate r-uniform hypergraphs on n vertices. This is tight up to the multiplicative constant and polylogarithmic term
What’s in the bin? Cities' waste data can guide smarter decisions
Get PDFAcross cities in Africa and Asia, waste management absorbs a huge share of municipal budgets but remains poorly understood. Insights from a forthcoming Cities that Work synthesis paper highlight how understanding the composition of waste can help city leaders select appropriate technologies, support informal recyclers, and design effective waste management policies for sustainable urban growth
Global, regional, and national burden of cardiovascular diseases and risk factors in 204 countries and territories, 1990-2023
Get PDFBackground: Cardiovascular diseases (CVDs) are the leading cause of mortality and are among the foremost causes of disability globally. CVD burden has continued to increase in most countries since 1990, with trends driven by changing exposures to harmful risk factors, population growth, and population aging. Objectives: We report estimates of global, national, and subnational CVD burden, including 18 subdiseases and 12 associated modifiable risk factors. We analyzed change in CVD burden from 1990 to 2023 and identified drivers of change including population growth, population aging, and risk factor exposure. Methods: The Global Burden of Disease (GBD) 2023 study, a multinational collaborative research study, quantified burden due to 375 diseases including CVD burden and identified drivers of change from 1990 to 2023 using all available data and statistical models. GBD 2023 estimated the population-level burden of diseases in 204 countries and territories from 1990 to 2023. Results: CVDs were the leading cause of disability-adjusted life years (DALYs) and deaths estimated in the GBD. As of 2023, there were 437 million (95% UI: 401 to 465 million) CVD DALYs globally, a 1.4-fold increase from the number in 1990 of 320 million (292 to 344 million). Ischemic heart disease, intracerebral hemorrhage, ischemic stroke, and hypertensive heart disease were the leading cardiovascular causes of DALYs in 2023 globally. As of 2023, age-standardized CVD DALY rates were highest in low and low-middle Socio-demographic Index (SDI) settings and lowest in high SDI settings. The number of CVD deaths increased globally from 13.1 million (95% UI: 12.2 to 14.0 million) in 1990 to 19.2 million (95% UI: 17.4 to 20.4 million) in 2023. The number of prevalent cases of CVD more than doubled since 1990, with 311 million (95% UI: 294 to 333 million) prevalent cases of CVD in 1990 and 626 million (95% UI: 591 to 672 million) prevalent cases in 2023 globally. A total of 79.6% (95% UI: 75.7% to 82.5%) of CVD burden is attributable to modifiable risk factors 347 million [95% UI: 318 to 373 million] DALYs in 2023). Globally, high systolic blood pressure, dietary risks, high low-density lipoprotein cholesterol, and air pollution were the modifiable risks responsible for most attributable CVD burden in 2023. Since 1990, changes in exposure to modifiable risk factors have had mixed effects on CVD burden, with increases in high body mass index, high fasting plasma glucose, and low physical activity leading to higher burden, while reductions in tobacco usage have mitigated some of these increases. Population growth and population aging were the main drivers of the increasing burden since 1990, adding 128 million (95% UI: 115 to 139 million) and 139 million (95% UI: 126 to 151 million) CVD DALYs to the increase in CVD burden since 1990. Conclusions: CVD remains the leading cause of disease burden and death worldwide with the greatest burden in low, low-middle, and middle SDI regions. Large variation exists in CVD burden even for countries at similar levels of development, a gap explained substantially by known, modifiable risk factors that are inadequately controlled. The decades-long increase in CVD burden was the result of population growth, population aging, and increased exposure to a subset of risk factors led by metabolic risks. Countries will need to adopt effective health system and public health strategies if they are to progress in achieving global goals to reduce the burden of CVD
Do multi-dimensional quotas improve social equality? Intersectional representation & group relations
No full textCan quotas mandating descriptive political representation effectively address social hierarchy? Quotas are typically utilized to resolve gender hierarchies that support women’s exclusion from political power. Yet social hierarchy is multi-dimensional. Mandating political inclusion on one identity (gender) may be insufficient. We posit that quotas mandating representation on two dimensions (ethnicity and gender) disrupt multi-dimensional hierarchy, improving inter-group relations beyond one-dimensional gender quotas. We analyze the causal effect of the world’s largest quota system, with quasi-random quotas for women, disadvantaged ethnicities, and women from disadvantaged ethnicities in India. Utilizing multiple datasets covering India since quota imposition, we find one-dimensional gender quotas temporarily lessen hierarchical barriers to intergroup interactions in public whereas one-dimensional ethnic quotas worsen interactions. However, two-dimensional quotas consistently diminish public and private hierarchy, durably improving inter-group relations. Suggestive evidence indicates this relationship travels globally. Our results highlight the opportunities, and limitations, of using descriptive representation to improve social relations
Reclaiming political agency: using theatre as resistance
Get PDFBased on a new version of Sophocles’ Antigone, Nikita argues that theatre can help activists move past the ‘narrative trap’ that portrays excluded groups as victims, depriving them of voice and agency
How air pollution is holding back India's economy
Get PDFEach year, smog descends on northern India. Schools close, work on construction sites halts and people change their routines in ways that undermine economic growth. India’s air pollution problem is often framed as a side effect of rapid development. Here Shefali Khanna presents the emerging body of economic evidence which suggests instead that polluted air is itself a significant obstacle to growth
Why is inequality so high in Latin America? An interview with Francisco H. G. Ferreira
Get PDFHave cash transfer programmes been successful in reducing poverty in Latin America? What does the region’s large “informal” sector mean for the adoption of universal policies (healthcare, childcare, basic income) to reduce inequality? Why is capital so highly concentrated – and to what extent is this shaped by historical legacies? And can Brazil balance economic development with protection of the Amazon rainforest? In this wide-ranging interview, Francisco H. G. Ferreira (Amartya Sen Professor of Inequality Studies at LSE and Director of the International Inequalities Institute) discusses these questions – and much more – in a stock-take of the state of inequality across Latin America. Asking the questions is Pranab Bardhan (Emeritus Professor at UC Berkeley)
Blow-up lemmas for sparse graphs
No full textThe blow-up lemma states that a system of super-regular pairs contains all bounded degree spanning graphs as subgraphs that embed into a corresponding system of complete pairs. This lemma has far-reaching applications in extremal combinatorics. We prove sparse analogues of the blow-up lemma for subgraphs of random and of pseudorandom graphs. Our main results are the following three sparse versions of the blow-up lemma: one for embedding spanning graphs with maximum degree ∆ in subgraphs of G(n, p) with p = C(logn/n) 1/∆ ; one for embedding spanning graphs with maximum degree ∆ and degeneracy D in subgraphs of G(n, p) with p = C (logn/n) 1/(2D+1) ; and one for embedding spanning graphs with maximum degree ∆ in (p, cpmax(4,(3∆+1)/2)n)-bijumbled graphs. We also consider various applications of these lemma
Audience evaluations of news videos made with various levels of automation: a population-based survey experiment
No full textThe use of automation in news content creation is expanding from the written to the audio-visual medium with news organizations including Reuters turning to video automation services provided by companies such as Wibbitz. Although researchers have explored audience perceptions of text-based news automation, to date no published study has examined how news consumers perceive automated news videos. We conducted a between-subjects online survey experiment to compare how a socio-demographically representative sample (n = 4200) of online news consumers in the UK perceived human-made, partly automated, and highly automated short-form online news videos on 14 different story topics. Our findings show that human-made videos received on average more favourable responses on some evaluation variables, although the differences were not large. We also found some significant differences in the relative evaluation of automated and human-made news videos across different individual stories. For practitioners our results suggest partly automated news videos with post-automation human editing can be well received. For researchers our results show the need to use reasonably large sets of experimental stimuli, and suggest that ensuring socio-demographic variation within samples of respondents is worthwhile
