Extreme Computing Research Center

Permanent URI for this collection

Browse

Recent Submissions

Now showing 1 - 5 of 625
  • Conference Paper

    TinyML with Meta-Learning on Microcontrollers for Air Pollution Prediction

    (MDPI, 2024-04-08) Wardana, I Nyoman Kusuma; Fahmy, Suhaib A.; Gardner, Julian W.; Computer Science; Computer Science Program; Extreme Computing Research Center; Computer, Electrical and Mathematical Sciences and Engineering; Computer, Electrical and Mathematical Science and Engineering (CEMSE) Division; School of Engineering, University of Warwick, Coventry CV4 7AL, UK; Department of Electrical Engineering, Politeknik Negeri Bali, Badung 80571, Indonesia

    Tiny machine learning (tinyML) involves the application of ML algorithms on resource-constrained devices such as microcontrollers. It is possible to improve tinyML performance by using a meta-learning approach. In this work, we proposed lightweight base models running on a microcontroller to predict air pollution and show how performance can be improved using a stacking ensemble meta-learning method. We used an air quality dataset for London. Deployed on a Raspberry Pi Pico microcontroller, the tinyML file sizes were 3012 bytes and 5076 bytes for the two base models we proposed. The stacked model could achieve RMSE improvements of up to 4.9% and 14.28% when predicting NO2 and PM2.5, respectively.

  • Preprint

    On the relevance of lift force modelling in turbulent wall flows with small inertial particles

    (arXiv, 2024-04-08) Gao, Wei; Shi, Pengyu; Parsani, Matteo; Costa, Pedro; Applied Mathematics and Computational Science, Computer Electrical and Mathematical Science and Engineering Division, Extreme Computing Research Center, King Abdullah University of Science and Technology, Thuwal 23955-6900, Saudi Arabia; Mechanical Engineering; Mechanical Engineering Program; Computer, Electrical and Mathematical Sciences and Engineering; Computer, Electrical and Mathematical Science and Engineering (CEMSE) Division; Physical Sciences and Engineering; Physical Science and Engineering (PSE) Division; Applied Mathematics and Computational Science; Applied Mathematics and Computational Science Program; Extreme Computing Research Center; Institut de Mecanique des Fluides de Toulouse (IMFT), Universit ´ e de Toulouse, CNRS, Toulouse, France; Helmholtz-Zentrum Dresden – Rossendorf, Institute of Fluid Dynamics, 01328 Dresden, Germany; Process & Energy Department, TU Delft, Leeghwaterstraat 39, 2628 CB, The Netherlands

    In particle-laden turbulent wall flows, lift forces can influence the near-wall turbulence. This has been recently observed in particle-resolved simulations, which, however, are too expensive to be used in upscaled models. Instead, point-particle simulations have been the method of choice to simulate the dynamics of these flows during the last decades. While this approach is simpler, cheaper, and physically sound for small inertial particles in turbulence, some issues remain. In the present work, we address challenges associated with lift force modelling in turbulent wall flows and the impact of lift forces in the near-wall flow. We performed direct numerical simulations (DNS) of small inertial point particles in turbulent channel flow for fixed Stokes number and mass loading while varying the particle size. Our results show that the particle dynamics in the buffer region, causing the apparent particle-to-fluid slip velocity to vanish, raise major challenges for accurately modelling lift forces. While our results confirm that lift forces have little influence on particle dynamics for sufficiently small particle sizes, for inner-scaled diameters of order one and beyond, lift forces become quite important near the wall. The different particle dynamics under lift forces result in the modulation of streamwise momentum transport in the near-wall region. We analyze this lift-induced turbulence modulation for different lift force models, and the results indicate that realistic models are critical for particle-modeled simulations to correctly predict turbulence modulation by particles in the near-wall region.

  • Conference Paper

    Efficient GPU-based Large MIMO Detection Algorithm for Next-Generation Communication Systems

    (IEEE, 2023-12-04) Dabah, Adel; Rezki, Z.; Ltaief, Hatem; Keyes, David E.; Alouini, Mohamed-Slim; King Abdullah University of Science and Technology, Computer Electrical and Mathematical Science and Engineering; Computer, Electrical and Mathematical Sciences and Engineering; Computer, Electrical and Mathematical Science and Engineering (CEMSE) Division; Extreme Computing Research Center; Applied Mathematics and Computational Science; Applied Mathematics and Computational Science Program; Office of the President; Electrical and Computer Engineering; Electrical and Computer Engineering Program; University of California Santa Cruz, USA

    Low latency and high throughput are critical features for 5G mobile communication systems and beyond, in which the support of large MIMO is essential. Signal detection in large Multiple-Input Multiple-Output (MIMO) is a paramount component of a communication system since its performance in terms of latency, error rate, and achieved throughput depends on it. In this paper, we demonstrate the ability of our proposed massively parallel non-linear detection approach to support a large number of antennas and sustain high throughput at the extreme low latency of next-generation mobile communication systems. Our proposed method operates on a search tree that models all possible combinations of the transmitted signal. It selects coefficients from different levels and navigates the tree toward the Maximum Likelihood (ML) solution. To maintain the low latency requirement, we leverage the significant computational power of the Graphics Processing Unit (GPU) by expressing operations in terms of matrix-matrix multiplications. The obtained results show the ability of our non-linear detection approach to deal with up to 120 antennas with one-millisecond latency while satisfying good error rate performance at a practical signal-to-noise ratio (SNR).

  • Preprint

    Multidimensional deconvolution with shared bases

    (arXiv, 2024-04-02) Sushnikova, Daria; Ravasi, Matteo; Keyes, David E.; Computer, Electrical and Mathematical Sciences and Engineering; Computer, Electrical and Mathematical Science and Engineering (CEMSE) Division; Earth Science and Engineering; Earth Science and Engineering Program; Physical Sciences and Engineering; Physical Science and Engineering (PSE) Division; Applied Mathematics and Computational Science; Applied Mathematics and Computational Science Program; Extreme Computing Research Center; Office of the President

    We address the estimation of seismic wavefields by means of Multidimensional Deconvolution (MDD) for various redatuming applications. While offering more accuracy than conventional correlation-based redatuming methods, MDD faces challenges due to the ill-posed nature of the underlying inverse problem and the requirement to handle large, dense, complex-valued matrices. These obstacles have long limited the adoption of MDD in the geophysical community. Recent interest in this technology has spurred the development of new strategies to enhance the robustness of the inversion process and reduce its computational overhead. We present a novel approach that extends the concept of block low-rank approximations, usually applied to linear operators, to simultaneously compress the operator, right-hand side, and unknowns. This technique greatly alleviates the data-heavy nature of MDD. Moreover, since in 3d applications the matrices do not lend themselves to global low rank approximations, we introduce a novel H2-like approximation. We aim to streamline MDD implementations, fostering efficiency and controlling accuracy in wavefield reconstruction. This innovation holds potential for broader applications in the geophysical domain, possibly revolutionizing the analysis of multi-dimensional seismic datasets.

  • Preprint

    Uniform-over-dimension convergence with application to location tests for high-dimensional data

    (arXiv, 2024-03-24) Chowdhury, Joydeep; Dutta, Subhajit; Genton, Marc G.; King Abdullah University of Science and Technology; Computer, Electrical and Mathematical Sciences and Engineering; Computer, Electrical and Mathematical Science and Engineering (CEMSE) Division; Statistics; Statistics Program; Extreme Computing Research Center; Indian Institute of Technology Kanpur

    Asymptotic methods for hypothesis testing in high-dimensional data usually require the dimension of the observations to increase to infinity, often with an additional condition on its rate of increase compared to the sample size. On the other hand, multivariate asymptotic methods are valid for fixed dimension only, and their practical implementations in hypothesis testing methodology typically require the sample size to be large compared to the dimension for yielding desirable results. However, in practical scenarios, it is usually not possible to determine whether the dimension of the data at hand conform to the conditions required for the validity of the high-dimensional asymptotic methods, or whether the sample size is large enough compared to the dimension of the data. In this work, a theory of asymptotic convergence is proposed, which holds uniformly over the dimension of the random vectors. This theory attempts to unify the asymptotic results for fixed-dimensional multivariate data and high-dimensional data, and accounts for the effect of the dimension of the data on the performance of the hypothesis testing procedures. The methodology developed based on this asymptotic theory can be applied to data of any dimension. An application of this theory is demonstrated in the two-sample test for the equality of locations. The test statistic proposed is unscaled by the sample covariance, similar to usual tests for high-dimensional data. Using simulated examples, it is demonstrated that the proposed test exhibits better performance compared to several popular tests in the literature for high-dimensional data. Further, it is demonstrated in simulated models that the proposed unscaled test performs better than the usual scaled two-sample tests for multivariate data, including the Hotelling's test for multivariate Gaussian data.