Background Water insecurity poses a significant global challenge to health and development. While the biophysical and economic impacts of inadequate water and sanitation are well documented, the complex emotional and social tolls of water insecurity are less understood— particularly in the global North. In this article, we advance understandings of the psychosocial dimensions of water insecurity in Detroit, MI, where an estimated 100 000 households have been disconnected from water and sanitation services since the city declared bankruptcy in 2013. Methods A community-based participatory research study was conducted among residents of a local food pantry. A culturally relevant measure of water insecurity was developed through ethnographic engagement, then administered alongside the Kessler Psychological Distress scale. Results Our models reveal a substantial, statistically significant effect of water insecurity on psychological distress. Additionally, financial stress in paying for water and sanitation produces significant distress, even independent of water supply status. Conclusions Curtailing water and sanitation access has complex, intersecting effects, including implications for community mental health. Rapidly rising utility rates across the USA, in the context of growing poverty, underscore the urgency of addressing this issue. The present study is the first we know of in the USA to examine the relationship between water insecurity and psychosocial distress.
Deep reinforcement learning has been successful in a variety of tasks, such as game playing and robotic manipulation. However, attempting to learn tabula rasa disregards the logical structure of many domains as well as the wealth of readily available knowledge from domain experts that could help "warm start" the learning process. We present a novel reinforcement learning technique that allows for intelligent initialization of a neural network weights and architecture. Our approach permits the encoding domain knowledge directly into a neural decision tree, and improves upon that knowledge with policy gradient updates. We empirically validate our approach on two OpenAI Gym tasks and two modified StarCraft 2 tasks, showing that our novel architecture outperforms multilayer-perceptron and recurrent architectures. Our knowledge-based framework finds superior policies compared to imitation learning-based and prior knowledge-based approaches. Importantly, we demonstrate that our approach can be used by untrained humans to initially provide >80% increase in expected reward relative to baselines prior to training (p < 0.001), which results in a >60% increase in expected reward after policy optimization (p = 0.011).
The ease of access to pre-trained transformers has enabled developers to leverage large-scale language models to build exciting applications for their users. While such pre-trained models offer convenient starting points for researchers and developers, there is little consideration for the societal biases captured within these model risking perpetuation of racial, gender, and other harmful biases when these models are deployed at scale. In this paper, we investigate gender and racial bias across ubiquitous pre-trained language models, including GPT-2, XLNet, BERT, RoBERTa, ALBERT and Dis-tilBERT. We evaluate bias within pre-trained transformers using three metrics: WEAT, sequence likelihood, and pronoun ranking. We conclude with an experiment demonstrating the ineffectiveness of word-embedding techniques, such as WEAT, signaling the need for more robust bias testing in transformers.
Recent events have focused attention on the perceived widening of the economic divide between urban and rural areas, and on the continued rise of national income inequality. We demonstrate that, in fact, the average income gap between urban (metropolitan) and rural (nonmetropolitan) households has not risen over the past 40 years, and makes virtually no contribution to national income inequality. Rising national inequality is driven by rising inequality within both urban and rural America, not by an urban/rural divergence. As is well known, the growing dispersion of household money income is partly driven by rising wage inequality, particularly in urban areas. Less well recognized is the role played by other income sources. We show that a decline in the progressivity of the distribution of social security payments and cash transfers, and an increase in the regressivity of the distribution of retirement incomes, have jointly made a comparably large contribution to rising income inequality. At the same time, the share of income from self‐employment has declined, particularly in rural America, and because self‐employment income is very unequally distributed, its diminution has retarded the growth of rural inequality. In 2014–15, however, rural inequality increased, cutting the urban/rural inequality gap in half.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.