30472 research outputs found
Sort by
Pandemic modeling, good and bad
Get PDFWhat kind of epidemiological modeling works well, and what kind doesn’t? This is determined by the nature of the target: the relevant causal relations are unstable across contexts, which tells against any modeling that assumes otherwise. I look at two influential examples from the Covid pandemic. The first example is the paper from Imperial College, London, that projected future infection rates under various policy scenarios, and that in March 2020 was influential in persuading the UK government to impose a lockdown (Ferguson et al 2020). Because it assumes stability, this first example of modeling fails: it carries no epistemic force at all. A different modeling strategy is required, one that is less ambitious but more effective. This is illustrated by the second example: the paper, also from Imperial College, London, that in December 2020 first estimated the transmissibility of the Alpha variant (Volz et al 2020). This second, contextual example of modeling works well
Automatic source code summarization with graph attention networks
Get PDFSource code summarization aims to generate concise descriptions for code snippets in a natural language, thereby facilitates program comprehension and software maintenance. In this paper, we propose a novel approach–GSCS–to automatically generate summaries for Java methods, which leverages both semantic and structural information of the code snippets. To this end, GSCS utilizes Graph Attention Networks to process the tokenized abstract syntax tree of the program, which employ a multi-head attention mechanism to learn node features in diverse representation sub-spaces, and aggregate features by assigning different weights to its neighbor nodes. GSCS further harnesses an additional RNN-based sequence model to obtain the semantic features and optimizes the structure by combining its output with a transformed embedding layer. We evaluate our approach on two widely-adopted Java datasets; the experiment results confirm that GSCS outperforms the state-of-the-art baselines
England and Wales
Get PDFThe Impact of COVID-19 on Prison Conditions and Penal Policy presents the results of a worldwide exchange of information on the impact of COVID-19 in prisons. It also focuses on the human rights questions that have been raised during the pandemic, relating to the treatment of prisoners in institutions for both juveniles and adults worldwide.
The prison population rate of England and Wales doubled between 1994 and 2010, remaining fairly stable after 2010. The prison estate remains blighted by poor material conditions and high levels of violence and self-harm. After years of under-investment and inadequate staffing, about half of the country’s prisons are overcrowded. Experts had feared that COVID-19 would take a heavy toll in prisons. By dint of highly restrictive regimes imposed across the estate over several months, the toll in lives and confirmed cases has been less heavy than was feared. This came at the cost of greatly reduced family and social contact, and even less access to education, work and rehabilitation programmes than before COVID-19 struck. Over the first ten months of the pandemic, the total prison population declined. This was due to the national lockdown and court closures, rather than the implementation of much-needed policies to reduce prisoner numbers
Qualitative differences in the spatiotemporal brain states supporting configural face processing emerge in adolescence in autism
Get PDFBackground: Studying the neural processing of faces can illuminate the mechanisms of compromised social expertise in autism. To resolve a longstanding debate, we examined whether differences in configural face processing in autism are underpinned by quantitative differences in the activation of typical face processing pathways, or the recruitment of non-typical neural systems.
Methods: We investigated spatial and temporal characteristics of event-related EEG responses to upright and inverted faces in a large sample of children, adolescents, and adults with and without autism. We examined topographic analyses of variance and global field power to identify group differences in the spatial and temporal response to face inversion. We then examined how quasi-stable spatiotemporal profiles – microstates – are modulated by face orientation and diagnostic group.
Results: Upright and inverted faces produced distinct profiles of topography and strength in the topographical analyses. These topographical profiles differed between diagnostic groups in adolescents, but not in children or adults. In the microstate analysis, the autistic group showed differences in the activation strength of normative microstates during early-stage processing at all ages, suggesting consistent quantitative differences in the operation of typical processing pathways; qualitative differences in microstate topographies during late-stage processing became prominent in adults, suggesting the increasing involvement of non-typical neural systems with processing time and over development.
Conclusions: These findings suggest that early difficulties with configural face processing may trigger later compensatory processes in autism that emerge in later development
'Economics in Law: Law in Economics': introduction to the special issue
Get PDFThis special issue presents a series of papers, each of which – in different ways – reflects upon the role of law in markets. Together, these papers throw light on the ever-evolving relationship between legal studies and the discipline of economics
A principal–agent approach for estimating firm efficiency: Revealing bank managerial behavior
Get PDFWe consider agency-theory-based structural and reduced form models of bank performance. In the structural models, we take into account the managerial decision-making processes, and reveal underlying managerial efforts and thereby managerial behavior. In an empirical application, we estimate performance of Eurozone banks using our novel structural and reduced form approaches by Markov Chain Monte Carlo techniques. Our findings show, for the first time, that bank underperformance persists in the Eurozone whereas considerable variability across Member States exists. Our agency-theory-based structural modeling would favor cooperation of all interested parties and towards higher financial integration
Synchronous vegetation response to the last glacial-interglacial transition in northwest Europe
Get PDFThe North Atlantic region experienced abrupt high-amplitude cooling at the onset of the Younger Dryas stadial. However, due to chronological uncertainties in the available terrestrial records it is unclear whether terrestrial ecosystem response to this event was instantaneous and spatially synchronous, or whether regional or time-transgressive lags existed. Here we use new palynological results from a robustly dated lake sediment sequence retrieved from lake Hämelsee (north Germany) to show that vegetation change started at 12,820 cal. yr BP, concurrent with the onset of changes in local climate. A comparison of the Hämelsee results to a compilation of precisely dated palynological records shows instant and, within decadal-scale dating uncertainty, synchronous response of the terrestrial plant community to Late-Glacial climate change across northwest Europe. The results indicate that the environmental impact of climate cooling was more severe than previously thought and illustrates the sensitivity of natural terrestrial ecosystems to external forcing
Coastal dune dynamics in embayed settings with sea-level rise – Examples from the exposed and macrotidal north coast of SW England
Get PDFCoastal dune systems are natural forms of coastal defence, but are expected to exhibit increased erosion rates due to climate change impacts, notably sea-level rise and, potentially, increased storminess. This is especially the case in embayed coastal settings, i.e., where there are no significant sediment inputs into the beach/dune system from longshore sources. Dune development is closely linked to that of beaches that lie seaward, but their temporal dynamics tend to be asynchronous. So, whereas beaches are generally highly variable over a short- to medium-term (event–decadal) time scales, potentially obscuring a longer-term (decadal–centennial) sea-level signal, dunes display a low-pass filtered response which may contain a sea-level signal.
In this study, we investigate the decadal-scale, inter-annual dynamics of 25 embayed coastal dune systems along the exposed and macrotidal north coast of SW England. We then compare the observed behaviour with that hindcasted from simple parametric models and forecast future dune retreat rates due to sea-level rise. We show that practically all exposed dune systems show retreat with a regionally-averaged retreat rate of the dune foot of 0.5 m yr−1. The majority of retreat occurred over a small number of especially energetic winters and it was found that dune retreat is not automatically linked to dune volumetric change. Many of the retreating dune systems display so called ‘dune roll-over’, characterised by removal of sediment from the dune face and deposition at the dune top. Observed dune retreat rates were 2–3 times larger than predicted using simple parametric retreat models forced by sea-level rise. This suggests that the retreat models are inappropriate and/or that sea-level rise in itself may be insufficient to explain the observed retreat and that increased winter storminess may be implicated.
A key factor in driving dune retreat is considered to be the number of hours that waves reach the dune foot or the excess runup energy present at the dune foot elevation. Both sea-level rise and enhanced storminess will increase exposure of the dune foot to energetic wave action and this is expected to accelerate dune retreat rates in these settings. Application of parametric shoreline retreat models that account for the acceleration in rate of sea-level rise predicts c. 40 m of dune retreat by 2100 with a considerable range in retreat (20–75 m), resulting from uncertainty in model choice and parameterisation. Simply extrapolating the current dune retreat rate also results in c. 40 m of dune retreat by 2100, but this approach ignores the potential acceleration in dune retreat rate due to an increase in the rate of sea-level rise.
The combination of analysis of multi-annual coastal dune morphological change along with application of dune retreat models can provide useful insights into future dune evolution for coastal planners and managers
The long history of the medical humanities
No full textAn accessible introduction to the Medical Humanities
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!