Search or filter publications

Filter by type:

Filter by publication type

Filter by year:

to

Results

  • Showing results for:
  • Reset all filters

Search results

  • Conference paper
    Valassakis P, Ding Z, Johns E, 2021,

    Crossing the gap: a deep dive into zero-shot sim-to-real transfer for dynamics

    , 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems, Publisher: IEEE

    Zero-shot sim-to-real transfer of tasks with complex dynamics is a highly challenging and unsolved problem. A number of solutions have been proposed in recent years, but we have found that many works do not present a thorough evaluation in the real world, or underplay the significant engineering effort and task-specific fine tuning that is required to achieve the published results. In this paper, we dive deeper into the sim-to-real transfer challenge, investigate why this issuch a difficult problem, and present objective evaluations of anumber of transfer methods across a range of real-world tasks.Surprisingly, we found that a method which simply injects random forces into the simulation performs just as well as more complex methods, such as those which randomise the simulator's dynamics parameters

  • Journal article
    Kuntz J, Thomas P, Stan G-B, Barahona Met al., 2021,

    Stationary distributions of continuous-time Markov chains: a review of theory and truncation-based approximations

    , SIAM Review, ISSN: 0036-1445

    Computing the stationary distributions of a continuous-time Markov chaininvolves solving a set of linear equations. In most cases of interest, thenumber of equations is infinite or too large, and cannot be solved analyticallyor numerically. Several approximation schemes overcome this issue by truncatingthe state space to a manageable size. In this review, we first give acomprehensive theoretical account of the stationary distributions and theirrelation to the long-term behaviour of the Markov chain, which is readilyaccessible to non-experts and free of irreducibility assumptions made instandard texts. We then review truncation-based approximation schemes payingparticular attention to their convergence and to the errors they introduce, andwe illustrate their performance with an example of a stochastic reactionnetwork of relevance in biology and chemistry. We conclude by elaborating oncomputational trade-offs associated with error control and some open questions.

  • Journal article
    Quilodrán-Casas C, Silva VS, Arcucci R, Heaney CE, Guo Y, Pain CCet al., 2021,

    Digital twins based on bidirectional LSTM and GAN for modelling COVID-19

    The outbreak of the coronavirus disease 2019 (COVID-19) has now spreadthroughout the globe infecting over 100 million people and causing the death ofover 2.2 million people. Thus, there is an urgent need to study the dynamics ofepidemiological models to gain a better understanding of how such diseasesspread. While epidemiological models can be computationally expensive, recentadvances in machine learning techniques have given rise to neural networks withthe ability to learn and predict complex dynamics at reduced computationalcosts. Here we introduce two digital twins of a SEIRS model applied to anidealised town. The SEIRS model has been modified to take account of spatialvariation and, where possible, the model parameters are based on official virusspreading data from the UK. We compare predictions from a data-correctedBidirectional Long Short-Term Memory network and a predictive GenerativeAdversarial Network. The predictions given by these two frameworks are accuratewhen compared to the original SEIRS model data. Additionally, these frameworksare data-agnostic and could be applied to towns, idealised or real, in the UKor in other countries. Also, more compartments could be included in the SEIRSmodel, in order to study more realistic epidemiological behaviour.

  • Journal article
    Espinosa-González AB, Delaney BC, Marti J, Darzi Aet al., 2021,

    The role of the state in financing and regulating primary care in Europe: a taxonomy

    , Health Policy, Vol: 125, Pages: 168-176, ISSN: 0168-8510

    Traditional health systems typologies were based on health system financing type, such as the well-known OECD typology. However, the number of dimensions captured in classifications increased to reflect health systems complexity. This study aims to develop a taxonomy of primary care (PC) systems based on the actors involved (state, societal and private) and mechanisms used in governance, financing and regulation, which conceptually represents the degree of decentralisation of functions. We use nonlinear canonical correlations analysis and agglomerative hierarchical clustering on data obtained from the European Observatory on Health Systems and Policy and informants from 24 WHO European Region countries. We obtain four clusters: 1) Bosnia Herzegovina, Czech Republic, Germany, Slovakia and Switzerland: corporatist and/or fragmented PC system, with state involvement in PC supply regulation, without gatekeeping; 2) Greece, Ireland, Israel, Malta, Sweden, and Ukraine: public and (re)centralised PC financing and regulation with private involvement, without gatekeeping; 3) Finland, Norway, Spain and United Kingdom: public financing and devolved regulation and organisation of PC, with gatekeeping; and 4) Bulgaria, Croatia, France, North Macedonia, Poland, Romania, Serbia, Slovenia and Turkey: public and deconcentrated with professional involvement in supply regulation, and gatekeeping. This taxonomy can serve as a framework for performance comparisons and a means to analyse the effect that different actors and levels of devolution or fragmentation of PC delivery may have in health outcomes.

  • Conference paper
    Amendola M, Arcucci R, Mottet L, Casas CQ, Fan S, Pain C, Linden P, Guo YKet al., 2021,

    Data Assimilation in the Latent Space of a Convolutional Autoencoder

    , Pages: 373-386, ISSN: 0302-9743

    Data Assimilation (DA) is a Bayesian inference that combines the state of a dynamical system with real data collected by instruments at a given time. The goal of DA is to improve the accuracy of the dynamic system making its result as real as possible. One of the most popular technique for DA is the Kalman Filter (KF). When the dynamic system refers to a real world application, the representation of the state of a physical system usually leads to a big data problem. For these problems, KF results computationally too expensive and mandates to use of reduced order modeling techniques. In this paper we proposed a new methodology we called Latent Assimilation (LA). It consists in performing the KF in the latent space obtained by an Autoencoder with non-linear encoder functions and non-linear decoder functions. In the latent space, the dynamic system is represented by a surrogate model built by a Recurrent Neural Network. In particular, an Long Short Term Memory (LSTM) network is used to train a function which emulates the dynamic system in the latent space. The data from the dynamic model and the real data coming from the instruments are both processed through the Autoencoder. We apply the methodology to a real test case and we show that the LA has a good performance both in accuracy and in efficiency.

  • Conference paper
    Afzali J, Casas CQ, Arcucci R, 2021,

    Latent GAN: Using a Latent Space-Based GAN for Rapid Forecasting of CFD Models

    , Pages: 360-372, ISSN: 0302-9743

    The focus of this study is to simulate realistic fluid flow, through Machine Learning techniques that could be utilised in real-time forecasting of urban air pollution. We propose a novel Latent GAN architecture which looks at combining an AutoEncoder with a Generative Adversarial Network to predict fluid flow at the proceeding timestep of a given input, whilst keeping computational costs low. This architecture is applied to tracer flows and velocity fields around an urban city. We present a pair of AutoEncoders capable of dimensionality reduction of 3 orders of magnitude. Further, we present a pair of Generator models capable of performing real-time forecasting of tracer flows and velocity fields. We demonstrate that the models, as well as the latent spaces generated, learn and retain meaningful physical features of the domain. Despite the domain of this project being that of computational fluid dynamics, the Latent GAN architecture is designed to be generalisable such that it can be applied to other dynamical systems.

  • Journal article
    Lertvittayakumjorn P, Toni F, 2021,

    Explanation-based human debugging of nlp models: a survey

    , Transactions of the Association for Computational Linguistics, Vol: 9, Pages: 1508-1528, ISSN: 2307-387X

    Debugging a machine learning model is hard since the bug usually involves the training data and the learning process. This becomes even harder for an opaque deep learning model if we have no clue about how the model actually works. In this survey, we review papers that exploit explanations to enable humans to give feedback and debug NLP models. We call this problem explanation-based human debugging (EBHD). In particular, we categorize and discuss existing work along three dimensions of EBHD (the bug context, the workflow, and the experimental setting), compile findings on how EBHD components affect the feedback providers, and highlight open problems that could be future research directions.

  • Journal article
    Nurek M, Rayner C, Freyer A, Taylor S, Järte L, MacDermott N, Delaney BCet al., 2021,

    Recommendations for the Recognition, Diagnosis, and Management of Patients with Post COVID-19 Condition ('Long COVID'): A Delphi Study

    , SSRN Electronic Journal
  • Journal article
    Arcucci R, Zhu J, Hu S, Guo Y-Ket al., 2021,

    Deep Data Assimilation: Integrating Deep Learning with Data Assimilation

    , APPLIED SCIENCES-BASEL, Vol: 11
  • Conference paper
    Paulino-Passos G, Toni F, 2021,

    Monotonicity and Noise-Tolerance in Case-Based Reasoning with Abstract Argumentation

    , Pages: 508-518
  • Conference paper
    Lauren S, Belardinelli F, Toni F, 2021,

    Aggregating Bipolar Opinions

    , 20th International Conference on Autonomous Agents and Multiagent Systems
  • Conference paper
    Rakicevic N, Cully A, Kormushev P, 2020,

    Policy manifold search for improving diversity-based neuroevolution

    , Publisher: arXiv

    Diversity-based approaches have recently gained popularity as an alternativeparadigm to performance-based policy search. A popular approach from thisfamily, Quality-Diversity (QD), maintains a collection of high-performingpolicies separated in the diversity-metric space, defined based on policies'rollout behaviours. When policies are parameterised as neural networks, i.e.Neuroevolution, QD tends to not scale well with parameter space dimensionality.Our hypothesis is that there exists a low-dimensional manifold embedded in thepolicy parameter space, containing a high density of diverse and feasiblepolicies. We propose a novel approach to diversity-based policy search viaNeuroevolution, that leverages learned latent representations of the policyparameters which capture the local structure of the data. Our approachiteratively collects policies according to the QD framework, in order to (i)build a collection of diverse policies, (ii) use it to learn a latentrepresentation of the policy parameters, (iii) perform policy search in thelearned latent space. We use the Jacobian of the inverse transformation(i.e.reconstruction function) to guide the search in the latent space. Thisensures that the generated samples remain in the high-density regions of theoriginal space, after reconstruction. We evaluate our contributions on threecontinuous control tasks in simulated environments, and compare todiversity-based baselines. The findings suggest that our approach yields a moreefficient and robust policy search process.

  • Journal article
    Ruiz LGB, Pegalajar MC, Arcucci R, Molina-Solana Met al., 2020,

    A time-series clustering methodology for knowledge extraction in energy consumption data

    , Expert Systems with Applications, Vol: 160, ISSN: 0957-4174

    In the Energy Efficiency field, the incorporation of intelligent systems in cities and buildings is motivated by the energy savings and pollution reduction that can be attained. To achieve this goal, energy modelling and a better understanding of how energy is consumed are fundamental factors. As a result, this study proposes a methodology for knowledge acquisition in energy-related data through Time-Series Clustering (TSC) techniques. In our experimentation, we utilize data from the buildings at the University of Granada (Spain) and compare several clustering methods to get the optimum model, in particular, we tested k-Means, k-Medoids, Hierarchical clustering and Gaussian Mixtures; as well as several algorithms to obtain the best grouping, such as PAM, CLARA, and two variants of Lloyd’s method, Small and Large. Thus, our methodology can provide non-trivial knowledge from raw energy data. In contrast to previous studies in this field, not only do we propose a clustering methodology to group time series straightforwardly, but we also present an automatic strategy to search and analyse energy periodicity in these series recursively so that we can deepen granularity and extract information at different levels of detail. The results show that k-Medoids with PAM is the best approach in virtually all cases, and the Squared Euclidean distance outperforms the rest of the metrics.

  • Journal article
    Mack J, Arcucci R, Molina-Solana M, Guo Y-Ket al., 2020,

    Attention-based Convolutional Autoencoders for 3D-Variational Data Assimilation

    , COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING, Vol: 372, ISSN: 0045-7825
  • Conference paper
    Kotonya N, Toni F, 2020,

    Explainable Automated Fact-Checking: A Survey

    , Barcelona. Spain, 28th International Conference on Computational Linguistics (COLING 2020), Publisher: International Committee on Computational Linguistics, Pages: 5430-5443

    A number of exciting advances have been made in automated fact-checkingthanks to increasingly larger datasets and more powerful systems, leading toimprovements in the complexity of claims which can be accurately fact-checked.However, despite these advances, there are still desirable functionalitiesmissing from the fact-checking pipeline. In this survey, we focus on theexplanation functionality -- that is fact-checking systems providing reasonsfor their predictions. We summarize existing methods for explaining thepredictions of fact-checking systems and we explore trends in this topic.Further, we consider what makes for good explanations in this specific domainthrough a comparative analysis of existing fact-checking explanations againstsome desirable properties. Finally, we propose further research directions forgenerating fact-checking explanations, and describe how these may lead toimprovements in the research area.v

  • Journal article
    Greenhalgh T, Thompson P, Weiringa S, Neves AL, Husain L, Dunlop M, Rushforth A, Nunan D, de Lusignan S, Delaney Bet al., 2020,

    What items should be included in an early warning score for remote assessment of suspected COVID-19? qualitative and Delphi study

    , BMJ Open, Vol: 10, Pages: 1-26, ISSN: 2044-6055

    Background To develop items for an early warning score (RECAP: REmote COVID-19 Assessment in Primary Care) for patients with suspected COVID-19 who need escalation to next level of care.Methods The study was based in UK primary healthcare. The mixed-methods design included rapid review, Delphi panel, interviews, focus groups and software development. Participants were 112 primary care clinicians and 50 patients recovered from COVID-19, recruited through social media, patient groups and snowballing. Using rapid literature review, we identified signs and symptoms which are commoner in severe COVID-19. Building a preliminary set of items from these, we ran four rounds of an online Delphi panel with 72 clinicians, the last incorporating fictional vignettes, collating data on R software. We refined the items iteratively in response to quantitative and qualitative feedback. Items in the penultimate round were checked against narrative interviews with 50 COVID-19 patients. We required, for each item, at least 80% clinician agreement on relevance, wording and cut-off values, and that the item addressed issues and concerns raised by patients. In focus groups, 40 clinicians suggested further refinements and discussed workability of the instrument in relation to local resources and care pathways. This informed design of an electronic template for primary care systems.Results The prevalidation RECAP-V0 comprises a red flag alert box and 10 assessment items: pulse, shortness of breath or respiratory rate, trajectory of breathlessness, pulse oximeter reading (with brief exercise test if appropriate) or symptoms suggestive of hypoxia, temperature or fever symptoms, duration of symptoms, muscle aches, new confusion, shielded list and known risk factors for poor outcome. It is not yet known how sensitive or specific it is.Conclusions Items on RECAP-V0 align strongly with published evidence, clinical judgement and patient experience. The validation phase of this study is ongoing.Tria

  • Conference paper
    Kotonya N, Toni F, 2020,

    Explainable Automated Fact-Checking for Public Health Claims

    , 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP(1) 2020), Publisher: ACL, Pages: 7740-7754

    Fact-checking is the task of verifying the veracity of claims by assessing their assertions against credible evidence. The vast major-ity of fact-checking studies focus exclusively on political claims. Very little research explores fact-checking for other topics, specifically subject matters for which expertise is required. We present the first study of explainable fact-checking for claims which require specific expertise. For our case study we choose the setting of public health. To support this case study we construct a new datasetPUBHEALTHof 11.8K claims accompanied by journalist crafted, gold standard explanations(i.e., judgments) to support the fact-check la-bels for claims1. We explore two tasks: veracity prediction and explanation generation. We also define and evaluate, with humans and computationally, three coherence properties of explanation quality. Our results indicate that,by training on in-domain data, gains can be made in explainable, automated fact-checking for claims which require specific expertise.

  • Journal article
    Casas CQ, Arcucci R, Wu P, Pain C, Guo Y-Ket al., 2020,

    A Reduced Order Deep Data Assimilation model

    , PHYSICA D-NONLINEAR PHENOMENA, Vol: 412, ISSN: 0167-2789
  • Journal article
    Wang S, Nadler P, Arcucci R, Yang X, Li L, Huang Y, Teng Z, Guo Yet al., 2020,

    A Bayesian Updating Scheme for Pandemics: Estimating the Infection Dynamics of COVID-19

    , IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE, Vol: 15, Pages: 23-33, ISSN: 1556-603X
  • Conference paper
    Liu S, Lin Z, Wang Y, Jianming Z, Perazzi F, Johns Eet al., 2020,

    Shape adaptor: a learnable resizing module

    , European Conference on Computer Vision 2020, Publisher: Springer Verlag, Pages: 661-677, ISSN: 0302-9743

    We present a novel resizing module for neural networks: shape adaptor, a drop-in enhancement built on top of traditional resizing layers, such as pooling, bilinear sampling, and strided convolution. Whilst traditional resizing layers have fixed and deterministic reshaping factors, our module allows for a learnable reshaping factor. Our implementation enables shape adaptors to be trained end-to-end without any additional supervision, through which network architectures can be optimised for each individual task, in a fully automated way. We performed experiments across seven image classification datasets, and results show that by simply using a set of our shape adaptors instead of the original resizing layers, performance increases consistently over human-designed networks, across all datasets. Additionally, we show the effectiveness of shape adaptors on two other applications: network compression and transfer learning.

This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.

Request URL: http://www.imperial.ac.uk:80/respub/WEB-INF/jsp/search-t4-html.jsp Request URI: /respub/WEB-INF/jsp/search-t4-html.jsp Query String: id=989&limit=20&page=8&respub-action=search.html Current Millis: 1711705023970 Current Time: Fri Mar 29 09:37:03 GMT 2024