检索结果-内蒙古大学图书馆

Online-compatible unsupervised nonresonant anomaly detection

学校读者我要写书评

暂无评论

Physical Review D 2022年第5期105卷 055006-055006页

作者： Vinicius Mikuni Benjamin Nachman David Shih National Energy Research Scientific Computing Center Berkeley Lab Berkeley California 94720 USA Physics Division Lawrence Berkeley National Laboratory Berkeley California 94720 USA Berkeley Institute for Data Science University of California Berkeley California 94720 USA NHETC Department of Physics & Astronomy Rutgers University Piscataway New Jersey 08854 USA

There is a growing need for anomaly detection methods that can broaden the search for new particles in a model-agnostic manner. Most proposals for new methods focus exclusively on signal sensitivity. However, it is not enough to select anomalous events—there must also be a strategy to provide context to the selected events. We propose the first complete strategy for unsupervised detection of nonresonant anomalies that includes both signal sensitivity and a data-driven method for background estimation. Our technique is built out of two simultaneously trained autoencoders that are forced to be decorrelated from each other. This method can be deployed off-line for nonresonant anomaly detection and is also the first complete on-line-compatible anomaly detection strategy. We show that our method achieves excellent performance on a variety of signals prepared for the ADC2021 data challenge.

关键词： Particle production Hypothetical particles Artificial neural networks Hadron colliders Machine learning

Exploring the exact limits of the real-time equation-of-motion coupled cluster cumulant Green’s functions

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Peng, Bo Pathak, Himadri Panyala, Ajay Vila, Fernando D. Rehr, John J. Kowalski, Karol Physical and Computational Science Directorate Pacific Northwest National Laboratory RichlandWA99354 United States Advanced Computing Mathematics and Data Division Pacific Northwest National Laboratory RichlandWA99354 United States Department of Physics University of Washington SeattleWA98195 United States

In this paper, we analyze the properties of the recently proposed real-time equation-of-motion coupled-cluster (RT-EOM-CC) cumulant Green’s function approach [J. Chem. Phys. 2020, 152, 174113]. We specifically focus on identifying the limitations of the original time-dependent coupled cluster (TDCC) ansatz and propose an enhanced double TDCC ansatz ensuring the exactness in the expansion limit. Additionally, we introduce a practical cluster-analysis-based approach for characterizing the peaks in the computed spectral function from the RT-EOM-CC cumulant Green’s function approach, which is particularly useful for the assignments of satellite peaks when many-body effects dominate the spectra. Our preliminary numerical tests focus on reproducing, approximating, and characterizing the exact impurity Green’s function of the three-site and four-site single impurity Anderson models using the RT-EOM-CC cumulant Green’s function approach. The numerical tests allow us to have a direct comparison between the RT-EOM-CC cumulant Green’s function approach and other Green’s function approaches in the numerical exact limit. © 2024, CC BY.

关键词： Green's function

Adaptively Lossy Image Compression for Onboard Processing

学校读者我要写书评

暂无评论

Adaptively Lossy Image Compression for Onboard Processing

IEEE Aerospace Conference

作者： Justin Goodwill David Wilson Sebastian Sabogal Alan D. George Christopher Wilson NSF Center for Space High-Performance and Resilient Computing (SHREC) University of Pittsburgh 4420 Bayard St Suite 560 Pittsburgh PA Software Engineering Division NASA Goddard Space Flight Center Science Data Processing Branch Code 587 8800 Greenbelt Rd Greenbelt MD

ISBN: (数字)9781728127347

ISBN: (纸本)9781728127354

More efficient image-compression codecs are an emerging requirement for spacecraft because increasingly complex, onboard image sensors can rapidly saturate downlink bandwidth of communication transceivers. While these codecs reduce transmitted data volume, many are compute-intensive and require rapid processing to sustain sensor data rates. Emerging next-generation small satellite (SmallSat) computers provide compelling computational capability to enable more onboard processing and compression than previously considered. For this research, we apply two compression algorithms for deployment on modern flight hardware: (1) end-to-end, neural-network-based, image compression (CNN-JPEG);and (2) adaptive image compression through feature-point detection (FPD-JPEG). These algorithms rely on intelligent data-processing pipelines that adapt to sensor data to compress it more effectively, ensuring efficient use of limited downlink bandwidths. The first algorithm, CNN-JPEG, employs a hybrid approach adapted from literature combining convolutional neural networks (CNNs) and JPEG;however, we modify and tune the training scheme for satellite imagery to account for observed training instabilities. This hybrid CNN-JPEG approach shows 23.5% better average peak signal-to-noise ratio (PSNR) and 33.5% better average structural similarity index (SSIM) versus standard JPEG on a dataset collected on the Space Test Program - Houston 5 (STP-H5-CSP) mission onboard the International Space Station (ISS). For our second algorithm, we developed a novel adaptive image-compression pipeline based upon JPEG that leverages the Oriented FAST and Rotated BRIEF (ORB) feature-point detection algorithm to adaptively tune the compression ratio to allow for a tradeoff between PSNR/SSIM and combined file size over a batch of STP-H5-CSP images. We achieve a less than 1% drop in average PSNR and SSIM while reducing the combined file size by 29.6% compared to JPEG using a static quality factor (QF) of 90.

关键词：

State-space models are accurate and efficient neural operators for dynamical systems

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Hu, Zheyuan Daryakenari, Nazanin Ahmadi Shen, Qianli Kawaguchi, Kenji Em Karniadakis, George Department of Computer Science National University of Singapore Singapore119077 Singapore Center for Biomedical Engineering School of Engineering Brown University ProvidenceRI02912 United States Division of Applied Mathematics Brown University ProvidenceRI02912 United States Advanced Computing Mathematics and Data Division Pacific Northwest National Laboratory RichlandWA United States

Physics-informed machine learning (PIML) has emerged as a promising alternative to classical methods for predicting dynamical systems, offering faster and more generalizable solutions. However, existing models, including recurrent neural networks (RNNs), transformers, and neural operators, face challenges such as long-time integration, long-range dependencies, chaotic dynamics, and extrapolation, to name a few. To this end, this paper introduces state-space models implemented in Mamba for accurate and efficient dynamical system operator learning. Mamba addresses the limitations of existing architectures by dynamically capturing long-range dependencies and enhancing computational efficiency through reparameterization techniques. To extensively test Mamba and compare against another 11 baselines, we introduce several strict extrapolation testbeds that go beyond the standard interpolation benchmarks. We demonstrate Mamba’s superior performance in both interpolation and challenging extrapolation tasks. Mamba consistently ranks among the top models while maintaining the lowest computational cost and exceptional extrapolation capabilities. Moreover, we demonstrate the good performance of Mamba for a real-world application in quantitative systems pharmacology for assessing the efficacy of drugs in tumor growth under limited data scenarios. Taken together, our findings highlight Mamba’s potential as a powerful tool for advancing scientific machine learning in dynamical systems modeling. (The code will be available at https://***/zheyuanhu01/State_ Space_Model_Neural_Operator upon acceptance.) © 2024, CC BY.

关键词： Recurrent neural networks

Improving solution accuracy and convergence for stochastic physics parameterizations with colored noise

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Stinis, Panos Lei, Huan Li, Jing Wan, Hui Advanced Computing Mathematics and Data Division Pacific Northwest National Laboratory RichlandWA99354 United States Atmospheric Sciences and Global Change Division Pacific Northwest National Laboratory RichlandWA99354 United States Department of Computational Mathematics Science and Engineering Department of Statistics and Probability Michigan State University East LansingMI48824 United States

Stochastic parameterizations are used in numerical weather prediction and climate modeling to help capture the uncertainty in the simulations and improve their statistical properties. Convergence issues can arise when time integration methods originally developed for deterministic differential equations are applied naively to stochastic problems. [7, 8] demonstrated that a correction term to various deterministic numerical schemes, known in stochastic analysis as the It^o correction, can help improve solution accuracy and ensure convergence to the physically relevant solution without substantial computational overhead. The usual formulation of the It^o correction is valid only when the stochasticity is represented by white noise. In this study, a generalized formulation of the It^o correction is derived for noises of any color. The formulation is applied to a test problem described by an advection-diffusion equation forced with a spectrum of fast processes. We present numerical results for cases with both constant and spatially varying advection velocities to show that, for the same time step sizes, the introduction of the generalized It^o correction helps to substantially reduce time integration error and significantly improve the convergence rate of the numerical solutions when the forcing term in the governing equation is rough (fast varying);alternatively, for the same target accuracy, the generalized It^o correction allows for the use of significantly longer time steps and hence helps to reduce the computational cost of the numerical *** Codes 65C30, 86A10 Copyright © 2019, The Authors. All rights reserved.

关键词： Advection

Efficient implementation of modern entropy stable and kinetic energy preserving discontinuous Galerkin methods for conservation laws

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Ranocha, Hendrik Schlottke-Lakemper, Michael Chan, Jesse Rueda-Ramírez, Andrés M. Winters, Andrew R. Hindenlang, Florian Gassner, Gregor J. Applied Mathematics University of Münster Germany High-Performance Computing Center Stuttgart University of Stuttgart Germany Computational and Applied Mathematics Rice University United States Department of Mathematics and Computer Science University of Cologne Germany Computational Mathematics Division of Applied Mathematics Linköping University Sweden Max Planck Institute for Plasma Physics NMPP division Garching Germany Department of Mathematics and Computer Science Center for Data and Simulation Science University of Cologne Germany

Many modern discontinuous Galerkin (DG) methods for conservation laws make use of summation by parts operators and flux differencing to achieve kinetic energy preservation or entropy stability. While these techniques increase the robustness of DG methods significantly, they are also computationally more demanding than standard weak form nodal DG methods. We present several implementation techniques to improve the efficiency of flux differencing DG methods that use tensor product quadrilateral or hexahedral elements, in 2D or 3D respectively. Focus is mostly given to CPUs and DG methods for the compressible Euler equations, although these techniques are generally also useful for GPU computing and other physical systems including the compressible Navier-Stokes and magnetohydrodynamics equations. We present results using two open source codes, *** written in Julia and FLUXO written in Fortran, to demonstrate that our proposed implementation techniques are applicable to different code bases and programming languages. Copyright © 2021, The Authors. All rights reserved.

关键词： Entropy

Score-Based Physics-Informed Neural Networks for High-Dimensional Fokker-Planck Equations

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Hu, Zheyuan Zhang, Zhongqiang Em Karniadakis, George Kawaguchi, Kenji Department of Computer Science National University of Singapore Singapore119077 Singapore Department of Mathematical Sciences Worcester Polytechnic Institute WorcesterMA01609 United States Division of Applied Mathematics Brown University USA ProvidenceRI02912 United States Advanced Computing Mathematics and Data Division Pacific Northwest National Laboratory RichlandWA United States

The Fokker-Planck (FP) equation is a foundational partial differential equation (PDE) in stochastic processes involving Brownian motions. However, the curse of dimensionality (CoD) poses a formidable challenge when dealing with high-dimensional FP equations. Although Monte Carlo simulation and (vanilla) Physics-Informed Neural Networks (PINNs) have shown the potential to tackle CoD, both methods exhibit significant numerical errors in high dimensions when dealing with the probability density function (PDF) associated with Brownian motion. The point-wise PDF values tend to decrease exponentially as dimensionality increases, surpassing the precision of numerical simulations and resulting in substantial errors. Moreover, due to its massive sampling, Monte Carlo fails to offer fast sampling. Modeling the logarithm likelihood (LL) via vanilla PINNs transforms the FP equation into a notoriously difficult Hamilton-Jacobi-Bellman (HJB) equation, which is impractical for PINN learning, whose error grows rapidly with dimension. To this end, we propose a novel approach utilizing a score-based solver to fit the score function in stochastic differential equations (SDEs). The score function, defined as the gradient of the LL, plays a fundamental role in inferring LL and PDF and enables fast SDE sampling, offering an effective means to overcome the CoD. Three fitting methods, Score Matching (SM), Sliced Score Matching (SSM), and Score-PINN, are introduced, each contributing unique advantages in computational complexity, accuracy, and generality. The proposed score-based SDE solver operates in two stages: first, employing score matching or Score-PINN to acquire the score function;and second, solving the LL via an ordinary differential equation (ODE) using the obtained score function. Comparative evaluations across these methods showcase varying trade-offs. The proposed methodology is evaluated across diverse SDEs, including anisotropic Ornstein-Uhlenbeck processes, geometric Browni

关键词： Fokker Planck equation

Environment scan of generative AI infrastructure for clinical and translational science

学校读者我要写书评

暂无评论

Npj health systems

Npj health systems 2025年第1期2卷 4页

作者： Betina Idnay Zihan Xu William G Adams Mohammad Adibuzzaman Nicholas R Anderson Neil Bahroos Douglas S Bell Cody Bumgardner Thomas Campion Mario Castro James J Cimino I Glenn Cohen David Dorr Peter L Elkin Jungwei W Fan Todd Ferris David J Foran David Hanauer Mike Hogarth Kun Huang Jayashree Kalpathy-Cramer Manoj Kandpal Niranjan S Karnik Avnish Katoch Albert M Lai Christophe G Lambert Lang Li Christopher Lindsell Jinze Liu Zhiyong Lu Yuan Luo Peter McGarvey Eneida A Mendonca Parsa Mirhaji Shawn Murphy John D Osborne Ioannis C Paschalidis Paul A Harris Fred Prior Nicholas J Shaheen Nawar Shara Ida Sim Umberto Tachinardi Lemuel R Waitman Rosalind J Wright Adrian H Zai Kai Zheng Sandra Soo-Jin Lee Bradley A Malin Karthik Natarajan W Nicholson Price Ii Rui Zhang Yiye Zhang Hua Xu Jiang Bian Chunhua Weng Yifan Peng Department of Biomedical Informatics Columbia University Irving Medical Center New York NY USA. Department of Population Health Sciences Weill Cornell Medicine New York NY USA. Department of Pediatrics Boston Medical Center Boston MA USA Chobanian & Avedisian School of Medicine Boston University Boston MA USA. Oregon Clinical and Translational Research Institute Oregon Health and Science University Portland OR USA. Department of Public Health Sciences University of California Davis Davis CA USA. Keck School of Medicine University of Southern California Los Angeles CA USA. Department of Medicine David Geffen School of Medicine University of California Los Angeles Los Angeles CA USA. Department of Pathology and Laboratory Medicine University of Kentucky College of Medicine Lexington KY USA. Clinical and Translational Science Center Weill Cornell Medicine New York NY USA. Division of Pulmonary Critical Care and Sleep Medicine University of Kansas School of Medicine Kansas City KS USA. Department of Biomedical Informatics and Data Science Heersink School of Medicine University of Alabama Birmingham AL USA. Harvard Law School Petrie-Flom Center for Health Law Policy Biotechnology and Bioethics Harvard University Cambridge MA USA. Department of Biomedical Informatics University at Buffalo Buffalo NY USA. Center for Clinical and Translational Science Mayo Clinic Rochester MN USA. Technology and Digital Solutions Stanford Medicine Stanford University Stanford CA USA. Center for Biomedical Informatics Rutgers Cancer Institute New Brunswick NJ USA. Department of Learning Health Sciences University of Michigan Medical School Ann Arbor MI USA. Altman Clinical and Translational Research Institute (ACTRI) University of California San Diego La Jolla CA USA. Department of Biostatistics and Health Data Science School of Medicine Indiana University Indianapolis IN USA. Department of Ophthalmology CCTSI University of Colorado Aurora CO USA. Center for Clinic

This study reports a comprehensive environmental scan of the generative AI (GenAI) infrastructure in the national network for clinical and translational science across 36 institutions supported by the CTSA Program led by the National Center for Advancing Translational sciences (NCATS) of the National Institutes of Health (NIH) at the United States. Key findings indicate a diverse range of institutional strategies, with most organizations in the experimental phase of GenAI deployment. The results underscore the need for a more coordinated approach to GenAI governance, emphasizing collaboration among senior leaders, clinicians, information technology staff, and researchers. Our analysis reveals that 53% of institutions identified data security as a primary concern, followed by lack of clinician trust (50%) and AI bias (44%), which must be addressed to ensure the ethical and effective implementation of GenAI technologies.

关键词： Health services Public health

Deep learning for cardiac image segmentation: A review

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Chen, Chen Qin, Chen Qiu, Huaqi Tarroni, Giacomo Duan, Jinming Bai, Wenjia Rueckert, Daniel Biomedical Image Analysis Group Department of Computing Imperial College London London United Kingdom Department of Computer Science City University of London London United Kingdom School of Computer Science University of Birmingham Birmingham United Kingdom Data Science Institute Imperial College London London United Kingdom Division of Brain Sciences Department of Medicine Imperial College London London United Kingdom

Deep learning has become the most widely used approach for cardiac image segmentation in recent years. In this paper, we provide a review of over 100 cardiac image segmentation papers using deep learning, which covers common imaging modalities including magnetic resonance imaging (MRI), computed tomography (CT), and ultrasound (US) and major anatomical structures of interest (ventricles, atria and vessels). In addition, a summary of publicly available cardiac image datasets and code repositories are included to provide a base for encouraging reproducible research. Finally, we discuss the challenges and limitations with current deep learning-based approaches (scarcity of labels, model generalizability across different domains, interpretability) and suggest potential directions for future research. Copyright © 2019, The Authors. All rights reserved.

关键词： Magnetic resonance imaging