13506 research outputs found
Sort by
Suppressing precision errors by connecting copies of Ising models for continuous-time quantum computing
Get PDFClassical optimization problems can be mapped to Ising models in order to be solved by continuous-time quantum computing. It was recognised in [Young et al, PRA (2013)], [Pearson et al, njp QI, (2019)] and [Albash et al, QST, (2019)], that these problems are susceptible to a lack of precision in the fields and couplings of the Ising model.
In this thesis we introduce a scheme first described in [Bennett et al, arXiv:2206.02545, (2022)], which aims to suppress errors caused by lack of precision. This scheme was inspired by quantum annealing correction (QAC), first introduced in [Young et al, PRA (2013)], where physical qubits in multiple copies of a Ising model are linked together.
However, we introduce several innovations thereby making our scheme distinct.
First, when determining the ground state of the problem, we require only one copy to be correct, because the solution quality can be checked efficiently.
Second, using this "one correct copy" setting, we find the optimal strength of links connecting the copies to be anti-ferromagnetic and close to the minimum strength allowed by the precision. Here we find an improvement (on average) above separate copies and copies connected ferromagnetically.
Third, we find that configurations of copies that contain frustration (e.g. a loop of three or five copies), provide a further improvement in fraction correct.
Numerically testing our innovations on small instances of spin chains and spin glasses, we find improved tolerance to lack of precision equating to around 3 bits of precision improvement at p=7.
We develop a link selection protocol which aims to determine in a computationally non-intensive fashion, whether or not to connect corresponding qubits in different copies. Here, we obtain mixed results, with improvement in fraction correct over separate copies only for precisions p<4.
Finally, we apply our error suppression scheme when computing with quantum walks. In this setting we find that the improvement from using our technique is lost for all values of precision. We hypothesise this is due to the way our error suppression scheme functions by allowing 'access' to excited states, available innately in quantum walks
The Nature of Supererogation and its Application in Medical Practice
Get PDFThis thesis emphasises the need for doctors' professional duties to be bound firmly together with their moral obligations; there is danger for the patient when there is a divergence. I believe that the consideration and practice of supererogation acts as a means to further this union. I argue for its inclusion in medical practice.
Acts of supererogation go beyond the requirements of duty. The concept has roots in Christianity, and was most fully developed by St Thomas Aquinas. It came under attack during the Reformation; for Protestants, salvation was not to be earned by good works, it was only bestowed by the grace of God.
The concept has largely disappeared from everyday usage in a secular society, nevertheless there is an expectation that members of the caring professions should embrace it to some extent. Doctors have lost an understanding of the subject, hampered by the professionalization of medicine.
This thesis illustrates the concept in its practical application and provides a framework of three different models of supererogation. I look to the work of K.E.Kirk and Stanley Hauerwas in theology, and I look at the philosophy of Emmanuel Levinas and Iris Murdoch to support my claims. Examples from literary sources, rather than medical case histories which are often too businesslike in tone, illustrate the complexity of what takes place between patient and doctor.
Doctors might fail to see what is important; they need the faculty of moral perception and also imagination to think of what more can be done for patients as well as wisdom to judge if this can be done safely. I believe that my third model of supererogation could be put into practice for the benefit of both patients and doctors
The Role of BrxR in Regulating Bacterial Phage-Defence Systems
Get PDFBacteria are under constant attack by bacteriophages, their natural predators that outnumber them 10-fold. The resulting selection pressure has given rise to a diverse range of phage-defence systems within bacteria. These systems are often clustered within ‘defence islands’ in the bacterial genome, facilitating coregulation and complementary action. In this thesis, regulation of an Escherichia fergusonii defence island containing both BREX and type IV restriction modification systems by BrxR protein is explored. Through LacZ assays, mutagenesis studies, and EOP assays, BrxR is functionally characterised as a ligand-binding transcriptional repressor of phage defence. The roles of the HTH and WYL domains found within BrxR are identified as likely DNA- and ligand-binding regions, respectively, and the groundwork is laid for future study of BrxR, including identification of its cognate phage-associated ligand
Percolative Current Flow through Anisotropic High-Field Superconductors under Strain
Get PDFFor decades, flux pinning scaling laws based upon unimodal, infinitesimally narrow, averaged distributions of critical superconducting parameters have been used to explain what limits the critical currents of practical superconducting materials in high fields. These scaling laws have enabled superconducting technologies ranging from MRI scanners, to high field research magnets, to magnets for particle accelerators and fusion energy. However in this work, we progress beyond these approximations and provide:
(i) The first comprehensive analysis of critical current density data showing that in high magnetic fields, technological
low and high temperature superconducting materials can be treated as a percolative network of Josephson junctions. We
then extract the size and normal state properties of the grain boundaries, the underlying distribution within the grains, and the dimensionality of the current flow within the material
(ii) The first reported measurements of the in-plane, biaxial strain dependence of for (RE)BCO tapes. This provides a description and understanding of the effects of the two most important strain components on and
(iii) The first comprehensive framework for percolative current flow in LTS and HTS superconductors under strain. It explicitly includes the factors suppressing and describes percolative flow within an anisotropic material containing a distribution of critical superconducting parameters. Our results show that large improvements to are available from further optimisation of the grains and grain boundaries in (RE)BCO and \ce{Nb3Sn} which will help enable the successful delivery of commercial fusion tokamaks
Socio-cultural factors that influence self-injury with suicidal intent in male prisoners
Get PDFUsing an intersectional lens, the aim of this study is to gain insight, and consider the application of the motivational and protective socio-cultural factors documented within service-user interviews as part of the Assessment, Care in Custody and Teamwork (ACCT) documentation. Firstly, this study utilises descriptive statistics to compare trends within HMP Bandwidth to National Statistics. This is followed by the main body of research: a reflexive, thematic analysis on six ACCT documents, taken from male prisoners who have expressed a desire or have physically engaged in self-injury within the general population of the reception prison. Findings revealed three themes that influenced the risk of self-injury within a male prison: adjusting to the physical prison regime, social factors of incarceration and distress surrounding medication. Findings are then discussed in detail through sub-themes, with reference to risk and protective factors, and practical recommendations. Conclusions indicate that socio-cultural factors and situational factors underlie the majority of motivational and protective factors for men who self-injure within prison, particularly Adverse Childhood Experiences and those incarcerated for the first-time
Low-cost household water treatment: A techno-behavioural intervention for local sustainable development in Afghanistan
Get PDFAccess to safe drinking water is a critical global challenge, in remote rural areas and urban centres alike. A pressing concern within this challenge lies in the sustainability of groundwater and the livelihoods reliant on it. However, a comprehensive study of such a complex issue as water insecurity requires a multidisciplinary approach that can synthesize perspectives from the natural and social sciences. With the overarching aim of studying and developing means to rectify water insecurity in low-income settings, this thesis pursues such an approach and contributes insights to the broader global dialogue through the case of the conflict-affected urban context of Kabul – where groundwater and livelihood challenges are driven especially by the contamination and rapid depletion of the local aquifers.
The multidisciplinary study begins with a geo-hydrology perspective that explores the sources of groundwater and the factors contributing to groundwater contamination. Additionally, it explores the potential of using clay disc filters for household water treatment from an earth sciences perspective. Complementing these natural science perspectives, the research also incorporates the COM-B framework, which draws from psychology and behavioural science. By leveraging anthropological techniques with a firm grounding in development research, the thesis further adopts a bottom-up approach to inform survey research.
Translating this multidisciplinary approach into the empirical research underlying this thesis, firstly, the groundwater recharge sources and groundwater dynamics in aquifers of Kabul city were explored relying on the analysis of the stable isotopic composition (δ18O and δ2H) of groundwater and surface water from the Upper Kabul River and Logar River. The results showed that precipitation was the primary source of recharge in the Central Kabul sub-basin, while mixed recharge from the river, precipitation, and irrigation return flow governed recharge in the Logar sub-basin. In the Paghman and Lower Kabul, and Upper Kabul sub-basins, increased rainfall input was also observed. The contribution of river water to groundwater recharge decreased from an average of over 60% in 2007 to less than 50% in 2020. Also, substantial groundwater level depletion was documented in the Central Kabul sub-basin and western parts of the city.
In addition to examining recharge sources and rates, the bacteriological and chemical characteristics of Kabul’s groundwater were analyzed. In Kabul, 4.1 million people rely on groundwater, making it critical to understand its contamination trends in the face of rapid development and social changes. The results showed an increase in E. coli and NO3-, indicating anthropogenic impacts on shallow groundwater quality. The Water Quality Index revealed that less than 35% of shallow groundwater samples had good quality. To address these issues, the implementation of point-of-use water purification was proposed as a temporary solution for reducing the occurrence of waterborne diseases.
Moreover, a qualitative study, based on 68 semi-structured interviews, explored the factors limiting access to clean drinking water in two peri-urban areas in Kabul. These factors included dysfunctional water supply networks, water price inequalities, uneven development, and aid prioritization. In addition, the stressors and dynamic access to water such as droughts, contamination, and electricity disruption were documented. Further, this research examined the nature and underlying factors of inter-household water-sharing practices. Water availability, the costs to the donor, the frequency of requests for water, the period over which they operate, and religious beliefs were all found to play key roles in determining water-sharing practices. The added influence of droughts in limiting water-sharing practices further highlighted the dynamics in performing the behaviour.
Furthermore, this research explored the factors that influence household water treatment practices, relied on a comprehensive behaviour change model (i.e., COM-B model). The results of the study showed that reflective and automatic motivation, as well as physical opportunity, had a statistically significant association with the performance of household water treatment behaviour. The findings suggest that socioeconomic, psychosocial, and contextual factors are all important in understanding and promoting household water treatment practices, and should be taken into account to develop interventions that are tailored to the specific needs and obstacles of different communities.
Lastly, the potential of using clay disc filters, frequently termed ceramic water filters, made from locally-sourced clay samples, was explored for removing bacteria from water. The clay discs were produced by mixing clay and sorted sawdust in a ratio of 1:2, and the filtration rate was 1 litre per hour. Clay disc filters have the potential to be a low-cost and locally-sourced solution for improving water quality in Afghanistan, but further research and development is needed to optimize their production, particularly by leveraging the skills of local potters in Kabul.
Overall, the synergistic combination of disciplinary techniques was thus capable of shedding light on the complex interplay between water resources, technology, and human behaviour (i.e., household water treatment) and provided a comprehensive understanding of the challenges and solutions surrounding access to safe drinking water
Natural Language Processing with Deep Latent Variable Models: Methods and Applications
Get PDFDue to their unparalleled performance and versatility, deep learning has become the de facto standard for building natural language processing (NLP) applications. Compared with conventional machine learning approaches, deep learning replaces extensive hand-engineered features in every task with end-to-end representation learning. Several concerns, however, have been raised in the research communities regarding their robustness, trustworthiness, explainability, and interpretability. Although these limitations of deep learning methods are widely acknowledged, work in methods and applications to alleviate these concerns in NLP is contrastingly limited. To address this research gap and explore a more robust approach for building NLP applications with deep learning, in this thesis, we studied deep latent variable models (DLVMs) in terms of methods (under supervised and semi-supervised learning settings) and applications (natural language understanding and generation) perspective for building natural language processing applications. We demonstrate the strength and benefits of DLVMs for NLP applications and discuss their effectiveness in addressing some of these concerns later in this thesis.
For contributions from a methods perspective, we studied the benefits of deep latent variable models in supervised and semi-supervised learning settings. These studies suggested that deep latent variable models are competitive in performance against standard deep learning methods; while offering additional robustness, trustworthiness, explainability and interoperability in various applications. For semi-supervised learning, particularly, we achieve state-of-the-art performance and prove the great potential of using deep latent variable models for semi-supervised learning problems.
For contributions from an applications perspective, we first presented two applications for language understanding problems, followed by two more applications for language generation problems. Our first application concerns a binary text classification task in the educational domain and pioneers the first research on how Bayesian deep learning can be applied to this text-based educational application. Our second application focuses on multilabel text classification tasks, and we present an efficient uncertainty quantification framework as our contribution. We demonstrate the effectiveness and generalisation of this framework with diverse architectures and present the first research on using deep latent variable models for efficient uncertainty quantification purposes in multilabel text classification tasks. Our third application deals with multiple explanation generation for an explainable artificial intelligence (XAI) task, and we present a first study on how deep latent variable models can be used to generate multiple explanations in the Stanford natural language inference task. In our last application, we explore paraphrase generation tasks and present the first study of DLVMs in a semi-supervised learning setting in paraphrase generation tasks; the DLVMs can enhance paraphrase generation performance when incorporating unlabelled data in a semi-supervised manner.
The findings in this thesis are of practical value to deep learning practitioners, researchers, and engineers working on a variety of problems in the field of natural language processing and deep learning
A microstructural and micromechanical investigation into shear dynamics during volcanic edifice collapse on Ascension Island: An experimental approach
Get PDFDuring gravitational collapse flows, shear forces are expressed through localised or
diffuse, brittle or ductile strain. Understanding material responses to shear within
gravitational collapse flows can be achieved through microstructural and
micromechanical investigation using established experimental techniques. This thesis
investigates a shear zone generated during a volcanic debris avalanche following the
collapse of a scoria cone on Green Mountain, Ascension Island through (1)
quantitative data on microstructural evolution within the shear zone through Scanning
Electron Microscope imaging, and (2) experimental work using rotary shear apparatus
to constrain the mechanical behaviour of the material under stress, and its influence
on internal microstructure. Microstructural analysis of the Green Mountain shear zone
reveals a decrease in grain size and porosity, as well as clast morphology evolution
toward the principal slip zone in the centremost region. Such observations are mirrored
in experimental shear zones presented herein. Mechanical data provide evidence that
material saturation promotes dynamic velocity weakening behaviour at seismic
velocities. Based on observations and evidence presented in this thesis, a model for
shear dynamics during the Green Mountain volcanic debris advance is proposed. It is
suggested that (1) a brittle cataclastic regime dominated within the shear zone,
resulting in the microstructural characteristics observed and (2) processes to facilitate
velocity weakening behaviour may include pore pressure fluidisation and nanoparticle
lubrication. Overall, this work contributes to the understanding of shear localisation,
internal microstructure, and facilitators of mechanical behaviour within the Green
Mountain volcanic debris avalanche deposit. Application of these findings to other
deposits and associated shear zones may help to better constrain collapse behaviour
and to mitigate associated risks
Impacts of sex, age, and desiccation on cuticular hydrocarbon profiles of Anopheles gambiae pupae
Get PDFMetal binding to the Polaris protein associated with ethylene sensing by plants
Get PDFCopper ions are essential to life, but toxic if not tightly regulated. In the model organism Arabidopsis thaliana, the ER-localised ethylene receptor, ETR1, requires Cu(I) at an intramembrane site, dependent on the Cu(I)-transporting P-type ATPase RAN1. However, the detailed biochemical mechanisms of Cu(I)-delivery, and ethylene binding, are unknown.
The protein Polaris (PLS), a negative regulator of ethylene signalling, shares some characteristics of known Cu(I)-metallochaperones, and was proposed to be involved in correct Cu(I)-metalation of ETR1. Here, metal binding to PLS has been investigated in-vitro, allowing prediction of its likely metalation state in-vivo. PLS bound Cu(I) and Zn(II) in 2:1 protein:metal stoichiometries, with β2 affinities of 3.79 x1019 and 3.76 x1012 M-2 respectively. Recently developed metalation calculators, based on metal-availability read-out from calibrated bacterial cells, were adapted to use these constants. The metal affinities of the Arabidopsis cytosolic Cu(I) chaperone Atx1, showed Cu(I) bound in a 1:1 and Zn(II) a 2:1 stoichiometry, and its metalation was modelled. This work showed, in E. coli BL21(DE3), by reading out CueR-dependent copA transcripts, Atx1 overexpression decreased Cu(I)-availability, when calibrated using E. coli JM109, with implications for heterologous expression of metalloproteins in bacteria. Availabilities, measured here, were used to correctly predict the metal preference of Atx1 in E. coli, when tested post-extraction. Using the Atx1 Cu(I)-affinity of 5.47 x10-18 M as an estimate for the intracellular buffered Cu(I)-availability in the cytosol of Arabidopsis, the metalation of PLS as a function of Atx1 Cu(I)-metalation showed it was unlikely PLS extracts Cu(I) directly from the buffer, at least not as a 2:1 complex.
This thesis speculates upon the putative roles of PLS in the biochemical activities of ETR1, and considers some of the implications and challenges associated with the potential formation of metal-dependent 2:1 ligand:metal complexes (with analogy to PLS2:Cu(I)) in biological systems, more broadly
Data sources:
Access Repository Dashboard
Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇