检索结果-内蒙古大学图书馆

Uncertainty-Informed Model Selection Method for Nonlinear System Identification and Interpretable Machine Learning

学校读者我要写书评

暂无评论

Uncertainty-Informed Model Selection Method for Nonlinear Sy...

Mediterranean Conference on Control and Automation (MED)

作者： Yuanlin Gu Hua-Liang Wei Division of Computer Science and Maths University of Stirling Stirling United Kingdom Department of Automatic Control and Systems Engineering University of Sheffield Sheffield

ISBN: (数字)9798350395440

ISBN: (纸本)9798350395457

关键词： Data-driven modeling Adaptation models Uncertainty Recurrent neural networks Training data Machine learning Data models

A Multi-Scale Feature Selection Framework for WiFi Access Points Line-of-sight Identification

学校读者我要写书评

暂无评论

A Multi-Scale Feature Selection Framework for WiFi Access Po...

IEEE Conference on Wireless Communications and Networking

作者： Xu Feng Khuong An Nguyen Zhiyuan Luo Computing & Maths Division University of Brighton Brighton United Kingdom Department of Computer Science Royal Holloway University of London Surrey United Kingdom

Despite its high accuracy in the ideal condition where there is a direct line-of-sight between the Access Points and the user, most WiFi indoor positioning systems struggle under the non-line-of-sight scenario. Thus, we propose a novel feature selection algorithm leveraging Machine Learning based weighting methods and multi-scale selection, with WiFi RTT and RSS as the input signals. We evaluate the algorithm performance on a campus building floor. The results indicated an accuracy of 93% line-of-sight detection success with 13 Access Points, using only 3 seconds of test samples at any moment; and an accuracy of 98% for individual AP line-of-sight detection.

关键词： Machine learning algorithms Buildings Line-of-sight propagation Machine learning Feature extraction Wireless fidelity Floors

Conformalised data synthesis

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Meister, Julia A. Nguyen, Khuong An Computing and Maths Division University of Brighton Lewes Road East Sussex BrightonBN2 4AT United Kingdom Department of Computer Science Royal Holloway University of London Egham Hill Egham SurreyTW20 0EX United Kingdom

With the proliferation of increasingly complicated Deep Learning architectures, data synthesis is a highly promising technique to address the demand of data-hungry models. However, reliably assessing the quality of a ‘synthesiser’ model’s output is an open research question with significant associated risks for high-stake domains. To address this challenge, we propose a unique synthesis algorithm that generates data from high-confidence feature space regions based on the Conformal Prediction framework. We support our proposed algorithm with a comprehensive exploration of the core parameter’s influence, an in-depth discussion of practical advice, and an extensive empirical evaluation of five benchmark datasets. To show our approach’s versatility on ubiquitous real-world challenges, the datasets were carefully selected for their variety of difficult characteristics: low sample count, class imbalance, and non-separability. In all trials, training sets extended with our confident synthesised data performed at least as well as the original set and frequently significantly improved Deep Learning performance by up to 61 percentage points *** Codes 68T37 © 2023, CC BY.

关键词： Data assimilation

A novel Deep Learning approach for one-step Conformal Prediction approximation

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Meister, Julia A. Nguyen, Khuong An Kapetanakis, Stelios Luo, Zhiyuan Computing and Maths Division University of Brighton East Sussex BrightonBN2 4GJ United Kingdom Distributed Analytics Solutions 17 Fawe Street LondonE14 6FD United Kingdom Department of Computer Science Royal Holloway University of London Egham SurreyTW20 0EX United Kingdom

Deep Learning predictions with measurable confidence are increasingly desirable for real-world problems, especially in high-risk settings. The Conformal Prediction (CP) framework is a versatile solution that guarantees a maximum error rate given minimal constraints [1]. In this paper, we propose a novel conformal loss function that approximates the traditionally two-step CP approach in a single step. By evaluating and penalising deviations from the stringent expected CP output distribution, a Deep Learning model may learn the direct relationship between the input data and the conformal p-values. We carry out a comprehensive empirical evaluation to show our novel loss function’s competitiveness for seven binary and multi-class prediction tasks on five benchmark datasets. On the same datasets, our approach achieves significant training time reductions up to 86% compared to Aggregated Conformal Prediction (ACP, [2]), while maintaining comparable approximate validity and predictive efficiency. © 2022, CC BY.

关键词： Forecasting

CochleaNet: A robust language-independent audio-visual model for speech enhancement

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Gogate, Mandar Dashtipour, Kia Adeel, Ahsan Hussain, Amir Edinburgh Napier University School of Computing EdinburghEH10 5DT United Kingdom University of Stirling Division of Computing Science and Maths StirlingFK9 4LA United Kingdom University of Wolverhampton School of Mathematics and Computer Science Wolverhampton United Kingdom

Noisy situations cause huge problems for suffers of hearing loss as hearing aids often make the signal more audible but do not always restore the intelligibility. In noisy settings, humans routinely exploit the audio-visual (AV) nature of the speech to selectively suppress the background noise and to focus on the target speaker. In this paper, we present a causal, language, noise and speaker independent AV deep neural network (DNN) architecture for speech enhancement (SE). The model exploits the noisy acoustic cues and noise robust visual cues to focus on the desired speaker and improve the speech intelligibility. To evaluate the proposed SE framework a first of its kind AV binaural speech corpus, called ASPIRE, is recorded in real noisy environments including cafeteria and restaurant. We demonstrate superior performance of our approach in terms of objective measures and subjective listening tests over the state-of-the-art SE approaches as well as recent DNN based SE models. In addition, our work challenges a popular belief that a scarcity of multi-language large vocabulary AV corpus and wide variety of noises is a major bottleneck to build a robust language, speaker and noise independent SE systems. We show that a model trained on synthetic mixture of Grid corpus (with 33 speakers and a small English vocabulary) and ChiME 3 Noises (consisting of only bus, pedestrian, cafeteria, and street noises) generalise well not only on large vocabulary corpora but also on completely unrelated languages (such as Mandarin), wide variety of speakers and noises. Copyright © 2019, The Authors. All rights reserved.

关键词： Audition

A Brain-Inspired Trust Management Model to Assure Security in a Cloud based IoT Framework for Neuroscience Applications

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Mahmud, Mufti Kaiser, M. Shamim Rahman, M. Mostafizur Rahman, M. Arifur Shabut, Antesar Al-Mamun, Shamim Hussain, Amir NeuroChip Lab University of Padova Padova35131 - Italy IIT Jahangirnagar University SavarDhaka1342 Bangladesh American International University - Bangladesh Dhaka1213 - Bangladesh Department of Computer Science University of Sheffield SheffieldS10 2TN United Kingdom Anglia Ruskin University CM1 1SQ - Chelmsford United Kingdom Saitama University Saitama338-8570 Japan Division of Computing Science & Maths University of Stirling StirlingFK9 4LA United Kingdom

Rapid popularity of Internet of Things (IoT) and cloud computing permits neuroscientists to collect multilevel and multichannel brain data to better understand brain functions, diagnose diseases, and devise treatments. To ensure secure and reliable data communication between end-to-end (E2E) devices supported by current IoT and cloud infrastructure, trust management is needed at the IoT and user ends. This paper introduces a Neuro-Fuzzy based Brain-inspired trust management model (TMM) to secure IoT devices and relay nodes, and to ensure data reliability. The proposed TMM utilizes node behavioral trust and data trust estimated using Adaptive Neuro-Fuzzy Inference System and weighted-additive methods respectively to assess the nodes trustworthiness. In contrast to the existing fuzzy based TMMs, the NS2 simulation results confirm the robustness and accuracy of the proposed TMM in identifying malicious nodes in the communication network. With the growing usage of cloud based IoT frameworks in Neuroscience research, integrating the proposed TMM into the existing infrastructure will assure secure and reliable data communication among the E2E devices. Copyright © 2018, The Authors. All rights reserved.

关键词： Internet of things

The dark energy survey supernova program: Cosmological biases from supernova photometric classification

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Vincenzi, Maria Sullivan, M. Möller, A. Armstrong, P. Bassett, B.A. Brout, D. Carollo, D. Carr, A. Davis, T.M. Frohmaier, C. Galbany, L. Glazebrook, K. Graur, O. Kelsey, L. Kessler, R. Kovacs, E. Lewis, G.F. Lidman, C. Malik, U. Nichol, R.C. Popovic, B. Sako, M. Scolnic, D. Smith, M. Taylor, G. Tucker, B.E. Wiseman, P. Aguena, M. Allam, S. Annis, J. Asorey, J. Bacon, D. Bertin, E. Brooks, D. Burke, D.L. Rosell, A. Carnero Carretero, J. Castander, F.J. Costanzi, M. da Costa, L.N. Pereira, M.E.S. De Vicente, J. Desai, S. Diehl, H.T. Doel, P. Everett, S. Ferrero, I. Flaugher, B. Fosalba, P. Frieman, J. García-Bellido, J. Gerdes, D.W. Gruen, D. Gutierrez, G. Hinton, S.R. Hollowood, D.L. Honscheid, K. James, D.J. Kuehn, K. Kuropatkin, N. Lahav, O. Li, T.S. Lima, M. Maia, M.A.G. Marshall, J.L. Miquel, R. Morgan, R. Ogando, R.L.C. Palmese, A. Paz-Chinchón, F. Pieres, A. Malagón, A.A. Plazas Reil, K. Roodman, A. Sanchez, E. Schubnell, M. Serrano, S. Sevilla-Noarbe, I. Suchyta, E. Tarle, G. To, C. Varga, T.N. Weller, J. Wilkinson, R.D. School of Physics and Astronomy University of Southampton SouthamptonSO17 1BJ United Kingdom Institute of Cosmology and Gravitation University of Portsmouth PortsmouthPO1 3FX United Kingdom Department of Physics Duke University DurhamNC27708 United States Centre for Astrophysics & Supercomputing Swinburne University of Technology VIC3122 Australia Université Clermont Auvergne CNRS IN2P3 LPC Clermont-FerrandF-63000 France The Research School of Astronomy and Astrophysics Australian National University ACT2601 Australia Madrid Spain African Institute for Mathematical Sciences 6 Melrose Road Muizenberg7945 South Africa Department of Maths and Applied Maths University of Cape Town Cape Town South Africa South African Astronomical Observatory Observatory Cape Town7925 South Africa Center for Astrophysics | Harvard & Smithsonian 60 Garden Street CambridgeMA02138 United States NASA Einstein INAF Astrophysical Observatory of Turin Pino TorineseI-10025 Italy School of Mathematics and Physics University of Queensland BrisbaneQLD4072 Australia Barcelona08034 Spain Campus UAB Carrer de Can Magrans s/n Barcelona08193 Spain Department of Astrophysics American Museum of Natural History New YorkNY United States Department of Astronomy and Astrophysics University of Chicago ChicagoIL60637 United States Kavli Institute for Cosmological Physics University of Chicago ChicagoIL60637 United States Argonne National Laboratory 9700 South Cass Avenue LemontIL60439 United States Sydney Institute for Astronomy School of Physics A28 The University of Sydney NSW2006 Australia Centre for Gravitational Astrophysics College of Science The Australian National University ACT2601 Australia Department of Physics and Astronomy University of Pennsylvania PhiladelphiaPA19104 United States Laboratório Interinstitucional de e-Astronomia - LIneA Rua Gal. José Cristino 77 RJ Rio de Janeiro20921-400 Brazil Fermi National Accelerator Laboratory P. O. B

Cosmological analyses of samples of photometrically-identified Type Ia supernovae (SNe Ia) depend on understanding the effects of 'contamination' from core-collapse and peculiar SN Ia events. We employ a rigorous analysis on state-of-the-art simulations of photometrically identified SN Ia samples and determine cosmological biases due to such 'non-Ia' contamination in the Dark Energy Survey (DES) 5-year SN sample. As part of the analysis, we test on our DES simulations the performance of SuperNNova, a photometric SN classifier based on recurrent neural networks. Depending on the choice of non-Ia SN models in both the simulated data sample and training sample, contamination ranges from 0.8-3.5 per cent, with the efficiency of the classification from 97.7-99.5 per cent. Using the Bayesian Estimation Applied to Multiple Species (BEAMS) framework and its extension BBC ('BEAMS with Bias Correction'), we produce a redshift-binned Hubble diagram marginalised over contamination and corrected for selection effects and we use it to constrain the dark energy equation-of-state, w. Assuming a flat universe with Gaussian ΩM prior of 0.311 ± 0.010, we show that biases on w are w,syst = 0.004. This compares to an expected statistical uncertainty of σw,stat = 0.039 for the DES-SN sample, thus showing that contamination is not a limiting uncertainty in our analysis. We also measure biases due to contamination on w0 and wa (assuming a flat universe), and find these to be 0 and a, hence 5 to 10 times smaller than the statistical uncertainties expected from the DES-SN sample. Copyright © 2021, The Authors. All rights reserved.

关键词： Surveys

Dark Energy Survey: A 2.1% measurement of the angular baryonic acoustic oscillation scale at redshift zeff=0.85 from the final dataset

学校读者我要写书评

暂无评论

Physical Review D 2024年第6期110卷 063515页

作者： T. M. C. Abbott M. Adamow M. Aguena S. Allam O. Alves A. Amon F. Andrade-Oliveira J. Asorey S. Avila D. Bacon K. Bechtol G. M. Bernstein E. Bertin J. Blazek S. Bocquet D. Brooks D. L. Burke H. Camacho A. Carnero Rosell D. Carollo A. Carr J. Carretero F. J. Castander R. Cawthon K. C. Chan C. Chang C. Conselice M. Costanzi M. Crocce L. N. da Costa M. E. S. Pereira T. M. Davis J. De Vicente N. Deiosso S. Desai H. T. Diehl S. Dodelson C. Doux A. Drlica-Wagner J. Elvin-Poole S. Everett I. Ferrero A. Ferté B. Flaugher P. Fosalba J. Frieman J. García-Bellido E. Gaztanaga G. Giannini K. Glazebrook R. A. Gruendl G. Gutierrez W. G. Hartley S. R. Hinton D. L. Hollowood K. Honscheid D. Huterer D. J. James S. Kent K. Kuehn O. Lahav S. Lee G. F. Lewis C. Lidman M. Lima H. Lin U. Malik C. Maraston J. L. Marshall P. Martini J. Mena-Fernández F. Menanteau R. Miquel J. J. Mohr J. Myles A. Möller R. C. Nichol R. L. C. Ogando A. Palmese W. J. Percival A. Pieres A. A. Plazas Malagón A. Porredon J. Prat M. Rodríguez-Monroy A. K. Romer A. Roodman R. Rosenfeld A. J. Ross E. S. Rykoff M. Sako S. Samuroff C. Sánchez E. Sanchez D. Sanchez Cid B. Santiago M. Schubnell I. Sevilla-Noarbe E. Sheldon M. Smith E. Suchyta M. E. C. Swanson G. Tarle D. Thomas C. To L. Toribio San Cipriano M. A. Troxel B. E. Tucker D. L. Tucker A. R. Walker N. Weaverdyck J. Weller P. Wiseman B. Yanny Cerro Tololo Inter-American Observatory NSF’s National Optical-Infrared Astronomy Research Laboratory Casilla 603 La Serena Chile Center for Astrophysical Surveys National Center for Supercomputing Applications 1205 West Clark Street Urbana Illinois 61801 USA Laboratório Interinstitucional de e-Astronomia—LIneA Rua Gal. José Cristino 77 Rio de Janeiro Rio de Janeiro-20921-400 Brazil Fermi National Accelerator Laboratory P. O. Box 500 Batavia Illinois 60510 USA Department of Physics University of Michigan Ann Arbor Michigan 48109 USA Institute of Astronomy University of Cambridge Madingley Road Cambridge CB3 0HA United Kingdom Kavli Institute for Cosmology University of Cambridge Madingley Road Cambridge CB3 0HA United Kingdom Department of Astrophysical Sciences Princeton University Peyton Hall Princeton New Jersey 08544 USA Departamento de Física Teórica and Instituto de Física de Partículas y del Cosmos (IPARCOS-UCM) Universidad Complutense de Madrid 28040 Madrid Spain Institut de Física d’Altes Energies (IFAE) The Barcelona Institute of Science and Technology Campus UAB 08193 Bellaterra (Barcelona) Spain Institute of Cosmology and Gravitation University of Portsmouth Portsmouth PO1 3FX United Kingdom Physics Department 2320 Chamberlin Hall University of Wisconsin-Madison 1150 University Avenue Madison Wisconsin 53706-1390 USA Department of Physics and Astronomy University of Pennsylvania Philadelphia Pennsylvania 19104 USA CNRS UMR 7095 Institut d’Astrophysique de Paris F-75014 Paris France Sorbonne Universités UPMC Univ Paris 06 UMR 7095 Institut d’Astrophysique de Paris F-75014 Paris France Department of Physics Northeastern University Boston Massachusetts 02115 USA University Observatory Faculty of Physics Ludwig-Maximilians-Universität Scheinerstraße 1 81679 Munich Germany Department of Physics and Astronomy University College London Gower Street London WC1E 6BT United Kingdom Kavli Institute for Particle Astrophysics and Cosmo

We present the angular diameter distance measurement obtained with the baryonic acoustic oscillation (BAO) feature from galaxy clustering in the completed Dark Energy Survey, consisting of six years (Y6) of observations. We use the Y6 BAO galaxy sample, optimized for BAO science in the redshift range 0.60.75 from any survey and the most precise measurement at any redshift from photometric surveys. The analysis was performed blinded to the BAO position, and it is shown to be robust against analysis choices, data removal, redshift calibrations, and observational systematics.

关键词： Cosmological parameters Cosmology Dark energy Sky surveys Large scale structure of the Universe

Sharing experiences with using next generation knowledge portals for advancing web communities

学校读者我要写书评

暂无评论

International Journal of Web Based Communities 2011年第3期7卷 256-275页

作者： Bessis, Nik Gaitan, Alfredo Shukla, Mitul Lai, Zhihua School of Computing and Maths University of Derby Kedleston Road Derby DE22 1GB United Kingdom Division of Psychology University of Bedfordshire Park Square LU1 3JU Luton United Kingdom Department of Computer Science and Technology University of Bedfordshire Park Square LU1 3JU Luton United Kingdom

In this article, we discuss the development of two repositories of shared knowledge using intra- and inter-tagging functionality. The functionality enables web user communities to upload, search, navigate through and tag multiple documents in multiple ways at a macro and micro level. Users can tag the whole document as well as parts there-in, as they deem appropriate. Our development process employed a unique, specifically designed community user needs approach (CUNA) to allow elicitation of diverse user communities needs. This included an iterative, rapid application development approach, which in turn, allowed lessons learned from the testing phase to be implemented in the following building phase. Results from in-house testing and an independent consultancy suggest that the web communities perceive the portals as beneficial not just during the project lifecycle but quite likely in future ventures. Copyright © 2011 Inderscience Enterprises Ltd.

关键词： Portals