30793 research outputs found
Sort by
Analyzing bi-objective optimization Pareto fronts using square shape slope index and NSGA-II: A multi-criteria decision-making approach
Get PDFData availability:
No data was used for the research described in the article.This paper introduces the Square Shape Slope Index (SSSI), a novel post-optimization multi-criteria decision-making (MCDM) approach for analyzing Pareto fronts generated from bi-objective optimization problems. SSSI leverages multiple Utopia and Nadir points—guided by a user-defined priority scale—to form a dynamic square region around particular segments of the Pareto front. Within this region, slope-based evaluations are used to rank solutions based on user preferences and criteria. The method’s effectiveness is demonstrated through empirical tests on diverse benchmark functions and real-world scenarios, such as energy distribution and portfolio optimization, each encompassing various shapes and patterns of the Pareto front. In addition, SSSI is compared against established decision-making approaches both geometrically and analytically using different aggregation methods. To account for the stochastic nature of evolutionary algorithms, the Non-Dominated Sorting Genetic Algorithm (NSGA-II) is employed to generate Pareto fronts for each test function. Results confirm the robustness and adaptability of SSSI, offering a clear and flexible framework for balancing conflicting objectives in multi-objective decision-making contexts
“Who am I? More than autism!” Race, Ethnicity, Culture and Autism
Get PDFThis thesis was submitted for the award of Doctor of Philosophy and was awarded by Brunel University LondonThe nexus between race, ethnicity, culture, identity and autism is a significantly under-researched area. Moreover, studies that explore the depth of Black British children and young people’s experiences and others at the intersection of special needs education in England is equally underexplored, despite large numbers of children and young people from Black Caribbean and African backgrounds (by birthright or ancestry) populating special schools. A crucial gap currently exists in English research that examines how children and young people with autism from Black backgrounds understand their racial, ethnic and cultural identities, and how these identities are influenced by home and school.
This research study is made up of two parts. The first involves auto/biographical accounts of my life as the mother of a young Black British man with autism, as well as my experiences as a special needs’ educator. The second part of this critical work is an investigation into the intersections of racial, ethnic and cultural identities among Black British children/ young people with autism attending English special schools. This qualitative study utilises data from over 40 semi-structured interviews to gather the first-hand experiences of those who can provide such insight and information to help better close this gap. They are Black British children/ young people with autism, their families and educators, exposing how the intersected characteristics of race, ethnicity and culture influence identity among children/ young people with autism in England.
Key findings from the study include Black British children/ young people with autism can understand their multiple identities; implicit and explicit ethnic-racial socialisation takes place predominately in the home; children and their families have experienced combined racism and ableism in special schools and other settings; and special school staff have experienced and witnessed racism in their provisions. Findings from this research have implications on special education in England
Applicability of the thermodynamic and mechanical route to Young’s equation for rigid and flexible solids: A molecular dynamics simulations study of a Lennard-Jones system model
Get PDFDATA AVAILABILITY:
The data that support the findings of this study are available within the article and its supplementary material available online at: https://pubs.aip.org/aip/jcp/article/162/5/054119/3333891/Applicability-of-the-thermodynamic-and-mechanical#90696696 .The wetting properties of a liquid in contact with a solid are commonly described by Young’s equation, which defines the relationship between the angle made by a fluid droplet onto the solid surface and the interfacial properties of the different interfaces involved. When modeling such interfacial systems, several assumptions are usually made to determine this angle of contact, such as a completely rigid solid or the use of the tension at the interface instead of the surface free energy. In this work, we perform molecular dynamics simulations of a Lennard-Jones liquid in contact with a Lennard-Jones crystal and compare the contact angles measured from a droplet simulation with those calculated using Young’s equation based on surface free energy or surface stress. We analyze cases where the solid atoms are kept frozen in their positions and where they are allowed to relax and simulate surfaces with different wettability and degrees of softness. Our results show that using either surface free energy or surface stress in Young’s equation leads to similar contact angles but different interfacial properties. We find that the approximation of keeping the solid atoms frozen must be done carefully, especially if the liquid can efficiently pack at the interface. Finally, we show that to correctly reproduce the measured contact angles when the solid becomes soft, the quantity to be used in Young’s equation is the surface free energy only and that the error committed in using the surface stress becomes larger as the softness of the solid increases.Acknowledgements: Computational facilities for this work were provided by the Computational Shared Facility (CSF) of the University of Manchester
Adaptive Resource Scheduling Algorithm for Multi-Feature Optimization in Personalized Wireless Body Area Networks
Get PDFWireless Body Area Network (WBANs), as a crucial technology in the field of healthcare monitoring, also plays a significant role in personal consumer electronics (CE). However, existing solutions have yet to effectively address challenges such as personalized demands, data heterogeneity, and dynamic link variations caused by diverse environmental factors. To address these issues, this study proposes a novel approach based on the IEEE 802.15.6 standard, which incorporates personalized node characteristics to cater to individual needs and adapt to node heterogeneity. Furthermore, a dynamic scheduling mechanism for node data is introduced, leveraging multi-feature environmental factors to enhance the practicality of WBAN systems. Finally, for real-time scheduling of emotional anomaly data, this study incorporates skin conductance nodes for the first time to evaluate emotional factors, ensuring the efficient transmission of urgent data. Theoretical analysis and simulation results demonstrate that the proposed approach significantly outperforms traditional methods in terms of energy efficiency, latency, throughput, and resource balance.10.13039/501100004502-Chongqing University of Posts and Telecommunications (Grant Number: BYJS202206);
10.13039/501100001809-National Natural Science Foundation of China (Grant Number: 62171073, 62206062, 62311530103 and U21A20447)
Extreme Wind Speed Prediction Under Noisy Labels: A Transfer-Learning-Assisted Cooperative Sample Selection Approach
No full textEGU General Assembly 2025 [Session NH9.5], Abstract EGU25-17880.Recently, deep learning (DL) techniques have been extensively applied to extreme weather prediction, which demonstrates their potential to address complex meteorological challenges. However, the success of DL-based weather prediction methods relies heavily on the availability of high-quality labelled training data. Human annotators and automated labelling tools may make mistakes due to limited expert knowledge or systematic errors, which leads to the noisy label problem. To address the noisy label challenge, we propose a novel transfer-learning-assisted cooperative sample selection (TLACSS) approach. A leader-follower cooperative learning strategy is put forward to mitigate the effects of noisy labels. To be specific, a leader network is first obtained based on transfer learning. Then, the leader network is jointly trained with two follower networks with the purpose of reducing the prediction divergence among the three networks. The small-loss criterion is employed to identify clean samples based on the joint loss function. A dynamic selection rate is introduced to automatically control the proportion of small-loss samples determined as clean during each epoch. The leader network, trained exclusively on the selected clean samples, is then utilized for extreme wind speed (EWS) prediction using real-world datasets. Furthermore, explainable artificial intelligence techniques are employed to improve the transparency and interpretability of the proposed TLACSS-based EWS prediction method
YOLO-ELWNet: A lightweight object detection network
Get PDFData availability:
Data will be made available on request.This paper proposes a YOLO-based efficient lightweight network (YOLO-ELWNet) for onboard object detection based on the YOLOv3. A channel split and shuffle with coordinate attention module is developed in the backbone block, which effectively reduces the size of model parameters and computational cost while maintaining the detection accuracy. A new feature fusion network is proposed in the neck block, where a cross-stage partial with efficient bottleneck module is put forward to improve the feature extraction ability and reduce the computational cost. The Scylla intersection over union-based loss function is utilized in the head block, which accelerates the convergence speed of the YOLO-ELWNet. The effectiveness of the proposed YOLO-ELWNet is validated on the open source KITTI vision benchmark. The performance of YOLO-ELWNet is superior to some mainstream lightweight object detection models in terms of detection accuracy and computational cost, which demonstrates its applicability for resource-constrained onboard object detection
Cubature Kalman Fusion Filtering Under Amplify- and-Forward Relays With Randomly Varying Channel Parameters
Get PDFIn this paper, the problem of cubature Kalman fusion filtering (CKFF) is addressed for multi-sensor systems under amplify-and-forward (AaF) relays. For the purpose of facilitating data transmission, AaF relays are utilized to regulate signal communication between sensors and filters. Here, the randomly varying channel parameters are represented by a set of stochastic variables whose occurring probabilities are permitted to exhibit bounded uncertainty. Employing the spherical-radial cubature principle, a local filter under AaF relays is initially constructed. This construction ensures and minimizes an upper bound of the filtering error covariance by designing an appropriate filter gain. Subsequently, the local filters are fused through the application of the covariance intersection fusion rule. Furthermore, the uniform boundedness of the filtering error covariance's upper bound is investigated through establishing certain sufficient conditions. The effectiveness of the proposed CKFF scheme is ultimately validated via a simulation experiment concentrating on a three-phase induction machine.10.13039/501100001809-National Natural Science Foundation of China (Grant Number: 12171124,61933007
The effects of meditation on attentional capacity, emotion regulation, and sensory information processing
Get PDFThis thesis was submitted for the award of Doctor of Philosophy and was awarded by Brunel University LondonEmerging evidence suggests beneficial effects of meditation generally, and mindfulness meditation specifically, on mental health and well-being via improved cognitive functioning and emotion regulation. The exact mechanisms underlying these effects, however, are yet to be fully understood, which is an overarching aim of three related studies reported in the thesis. One possible mechanism facilitating both more efficient attentional processing and emotion regulation in meditators is reduced attentional capture by salient stimuli, quantified in the present research using Attentional Blink (AB) phenomenon - a temporary inability to consciously perceive an attended stimulus when it is shortly preceded by another. Previous research using AB paradigms with neutral and emotional stimuli (referred to as Neutral Attention Blink (NAB) and Emotional Attentional Blink (EAB) paradigms, respectively) has shown that meditation practice (particularly mindfulness as secularly defined) attenuates NAB. Lower NAB has also been associated with higher trait non-reactivity (an aspect of mindfulness), both dispositional (innate) and trained through meditation practice.
Study 1 (online behavioural study) investigated the effects of meditation on AB using both the NAB and EAB paradigms in meditators (n = 75) and non-meditators (n = 54), as well as the relationships of NAB and EAB magnitude with each other and with trait non-reactivity and equanimity (non-attachment), with the latter being another aspect of mindful awareness that should attenuate attentional capture by salient (emotional) stimuli. There were no significant NAB or EAB differences between meditators and non-meditators. However, lower EAB was significantly associated with higher trait non-reactivity and equanimity in meditators as a group and in the meditator practicing mindfulness as secularly defined, suggesting that non-reactivity and equanimity might be attenuating EAB via reduced attentional capture by emotional stimuli.
Another mechanism that has been proposed to underlie AB is ‘gating’ or filtering of the sensory stimuli as a way to protect the processing of the first stimulus, given limited attentional capacity resources. Sensory gating is a fundamental mechanism protecting limited attentional processing capacity by an automatic inhibition of subsequent stimuli whilst the current stimulus is being processed and has been extensively studied using Prepulse Inhibition (PPI) – a modulation of the startle reflex when a startling stimulus (pulse) is preceded by a stimulus of less intensity (prepulse). Study 2 (lab-based study) explored the relationships of PPI with AB (with the focus on NAB) as well as with trait non-reactivity and equanimity in a separate sample of meditators (n = 23) and non-meditators (n = 21). There were no significant PPI differences between meditators and non-meditators and no significant associations of PPI with trait non-reactivity or equanimity. However, higher PPI was significantly associated with slower reaction times during NAB paradigm performance. The implications of this finding for the shared (or similar) mechanisms underlying AB and PPI are not clear, but given that PPI is an index of sensorimotor gating, they point to the relationship between automated (PPI) and voluntary (NAB) motor aspects engaged by the two paradigms. Yet another mechanism by which non-reactivity and equanimity might exert their attenuating effect on AB is the reduced activation of associative semantic memory networks in response to conceptually meaningful and/or emotionally salient stimuli, which has been observed in meditators in previous research. Associative semantic memory networks activation, alongside attentional capture, has been proposed as one of the mechanisms underlying the Affective Priming (AP) phenomenon – an impact that a first stimulus (prime) has on the speed and evaluation of the subsequent stimulus (target). Study 3 (online behavioural study) investigated the effects of meditation on AP and its relationship with EAB as well as trait non-reactivity and equanimity in meditators (n = 49) and non-meditators (n = 55) (a subsample of Study 1 participants). Congruent (same valence of primes and targets) and incongruent (opposite valence of primes and targets) were used in the AP paradigm. As expected, meditators were less impacted by the emotional valence of the primes or their congruity/incongruity with the targets, whereas non-meditators (particularly females) showed a strong priming effect. The correlation pattern between AP and EAB was somewhat inconsistent. Lower AP was associated with higher trait non-reactivity and equanimity in meditators for incongruent condition, suggesting that smaller AP in meditators may be due to reduced activation by the primes of associative semantic networks.
Together, the findings highlight trait non-reactivity and equanimity as the mechanisms underlying the effects of meditation generally, and mindfulness meditation specifically, on attentional capacity and emotion regulation via reduced attentional capture and semantic associative network activation by affective stimuli. Longitudinal studies using the AB, PPI, and AP paradigms are required to further investigate where the observed cross-sectional and correlational findings are the effects of meditation rather than pre-existing individual differences between meditators and non-meditations (and within meditators) with the view of developing particularly AB and AP paradigms as objective tools of meditation practice on attentional capacity and emotion regulation
Reweighting simulated events using machine-learning techniques in the CMS experiment
Get PDFData Availability Statement:
The manuscript has no associated data CERN for the benefit of the CMS Collaboration. [Author’s comment: Release and preservation of data used by the CMS Collaboration as the basis for publications is guided by the CMS data preservation, re-use and open access policy.]Code Availability Statement:
The manuscript has associated code/software in a data repository. [Author’s comment: The CMS core software is publicly available on GitHub (https://github.com/cms-sw/cmssw) policy (https://doi.org/10.7483/OPENDATA.CMS.1BNU.8V1W)].Data analyses in particle physics rely on an accurate simulation of particle collisions and a detailed simulation of detector effects to extract physics knowledge from the recorded data. Event generators together with a geant-based simulation of the detectors are used to produce large samples of simulated events for analysis by the LHC experiments. These simulations come at a high computational cost, where the detector simulation and reconstruction algorithms have the largest CPU demands. This article describes how machine-learning (ML) techniques are used to reweight simulated samples obtained with a given set of parameters to samples with different parameters or samples obtained from entirely different simulation programs. The ML reweighting method avoids the need for simulating the detector response multiple times by incorporating the relevant information in a single sample through event weights. Results are presented for reweighting to model variations and higher-order calculations in simulated top quark pair production at the LHC. This ML-based reweighting is an important element of the future computing model of the CMS experiment and will facilitate precision measurements at the High-Luminosity LHC.Funded by SCOAP3
Invisible in life and death: Visibilising the deaths of female Nepali migrant domestic workers and the struggles of their families
Get PDFName of Contributors: Sulochana Khanal, Pooja Bishwakarma, Sangita Nepali, Dil Kumari Budha Magar, Rusha Bhandari, Kalpana Chaudhary, Kamala Khadka Budha Magar, Srijana Timilsina, Chandika Mohara, Anupma Pokharel, and Sunita MainaliCollaborating Institutions: Brunel University of London, WOREC, and Sunita FoundationExecutive Summary In this report, we offer an urgent examination of the deaths of female Nepali migrant domestic workers abroad and the consequences for their families. Drawing on 17 cases across multiple districts in Nepal, we foreground the voices of bereaved family members and argue that Nepal’s migration industry systemically fails its most vulnerable citizens – women from Dalit, janajati, and low-income backgrounds. At the heart of this pilot feminist research is an important question: What happens when the state that benefits from migrant women’s labour disowns them in death?Brunel University of London via Brunel Research Initiative and Enterprise Fund (BRIEF Award) – 2024-25
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! 👇