检索结果-内蒙古大学图书馆

Unsupervised learning of atomic environments from simple features

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Reinhart, Wesley F. Department of Materials Science and Engineering Pennsylvania State University University ParkPA16802 United States Institute for Computational and Data Sciences Pennsylvania State University University ParkPA16802 United States

I present a strategy for unsupervised manifold learning on local atomic environments in molecular simulations based on simple rotation- and permutation-invariant three-body features. These features are highly descriptive, generalize to multiple chemical species, and are human-interpretable. The low-dimensional embeddings of each atomic environment can be used to understand and quantify messy crystal structures such as those near interfaces and defects or well-ordered crystal lattices such as in bulk materials without modification. The same method can also yield collective variables describing collections of particles such as for an entire simulation domain. I demonstrate the method on colloidal crystallization, ice crystals, and binary mesophases to illustrate its broad applicability. In each case, the learned latent space yields insights into the details of the observed microstructures. For ices and mesophases, supervised classifiers are trained based on the learned manifolds and directly compared against a recent neuralnetwork- based approach. Notably, while this method provides comparable classification performance, it can also be deployed on even a handful of observed environments without labels or a priori knowledge. Thus, the current approach provides an incredibly versatile strategy to characterize and classify local atomic environments, and may unlock insights in a wide variety of molecular simulation contexts. Copyright © 2021, The Authors. All rights reserved.

关键词： Molecular structure

Differentiable Programming for Differential Equations: A Review

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Sapienza, Facundo Bolibar, Jordi Schäfer, Frank Groenke, Brian Pal, Avik Boussange, Victor Heimbach, Patrick Hooker, Giles Pérez, Fernando Persson, Per-Olof Rackauckas, Christopher Department of Statistics University of California Berkeley United States Univ. Grenoble Alpes CNRS IRD G-INP Institut des Géosciences de l’Environnement Grenoble France TU Delft Department of Geosciences and Civil Engineering Delft Netherlands CSAIL Massachusetts Institute of Technology Cambridge United States TU Berlin Department of Electrical and Computer Engineering Berlin Germany Helmholtz Centre for Environmental Research Leipzig Germany Swiss Federal Research Institute WSL Birmensdorf Switzerland Oden Institute for Computational Engineering and Sciences University of Texas Austin United States Jackson School of Geosciences University of Texas Austin United States Department of Statistics and Data Science University of Pennsylvania United States Department of Mathematics University of California Berkeley United States Massachusetts Institute of Technology Cambridge United States JuliaHub Cambridge United States

The differentiable programming paradigm is a cornerstone of modern scientific computing. It refers to numerical methods for computing the gradient of a numerical model’s output. Many scientific models are based on differential equations, where differentiable programming plays a crucial role in calculating model sensitivities, inverting model parameters, and training hybrid models that combine differential equations with data-driven approaches. Furthermore, recognizing the strong synergies between inverse methods and machine learning offers the opportunity to establish a coherent framework applicable to both fields. Differentiating functions based on the numerical solution of differential equations is nontrivial. Numerous methods based on a wide variety of paradigms have been proposed in the literature, each with pros and cons specific to the type of problem investigated. Here, we provide a comprehensive review of existing techniques to compute derivatives of numerical solutions of differential equations. We first discuss the importance of gradients of solutions of differential equations in a variety of scientific domains. Second, we lay out the mathematical foundations of the various approaches and compare them with each other. Third, we cover the computational considerations and explore the solutions available in modern scientific software. Last but not least, we provide best-practices and recommendations for practitioners. We hope that this work accelerates the fusion of scientific models and data, and fosters a modern approach to scientific *** Codes 34-04, 49K40, 65D25, 65L09, 65M32, 86A22, 90C31 © 2024, CC BY.

关键词： Differential equations

Finite-size effects on the convergence time in continuous-opinion dynamics

学校读者我要写书评

暂无评论

Physical Review E 2021年第1期104卷 014309-014309页

作者： Hang-Hyun Jo Naoki Masuda Department of Physics The Catholic University of Korea Bucheon 14662 Republic of Korea Department of Mathematics State University of New York at Buffalo Buffalo New York 14260-2900 USA Computational and Data-Enabled Science and Engineering Program State University of New York at Buffalo Buffalo New York 14260-5030 USA

We study finite-size effects on the convergence time in a continuous-opinion dynamics model. In the model, each individual's opinion is represented by a real number on a finite interval, e.g., [0,1], and a uniformly randomly chosen individual updates its opinion by partially mimicking the opinion of a uniformly randomly chosen neighbor. We numerically find that the characteristic time to the convergence increases as the system size increases according to a particular functional form in the case of lattice networks. In contrast, unless the individuals perfectly copy the opinion of their neighbors in each opinion updating, the convergence time is approximately independent of the system size in the case of regular random graphs, uncorrelated scale-free networks, and complete graphs. We also provide a mean-field analysis of the model to understand the case of the complete graph.

关键词： Complex systems Nonequilibrium statistical mechanics Social systems Homogeneous networks Scale free & inhomogeneous networks Social dynamics Master equation Monte Carlo methods Network Models

Exploiting the Intrinsic Neighborhood Semantic Structure for Domain Adaptation in EEG-based Emotion Recognition

学校读者我要写书评

暂无评论

IEEE Transactions on Affective Computing 2025年

作者： Yang, Yi Wang, Ze Song, Yu Jia, Ziyu Wang, Boyu Jung, Tzyy-Ping Wan, Feng Macau University of Science and Technology Macao Centre for Mathematical Sciences Respiratory Disease AI Laboratory on Epidemic Intelligence and Medical Big Data Instrument Applications Faculty of Innovation Engineering 999078 China Tianjin University of Technology School of Electrical Engineering and Automation Tianjin Key Laboratory of New Energy Power Conversion Transmission and Intelligent Control Tianjin300384 China Chinese Academy of Sciences Beijing Key Laboratory of Brainnetome and Brain-Computer Interface and Brainnetome Center Institute of Automation Beijing100045 China Western University Department of Computer Science Brain Mind Institute LondonONN6A 3K7 Canada University of California at San Diego Swartz Center for Computational Neuroscience Institute for Neural Computation La Jolla CA92093 United States University of Macau Department of Electrical and Computer Engineering Faculty of Science and Technology China University of Macau Centre for Cognitive and Brain Sciences Centre for Artificial Intelligence and Robotics Institute of Collaborative Innovation 999078 China

Due to the inherent non-stationarity and individual differences present in electroencephalogram (EEG) signals, developing a generalizable model that performs well on new subjects is challenging in EEG-based emotion recognition. Most existing domain adaptation (DA) methods typically mitigate these discrepancies by aligning the marginal distributions of domain feature representations. However, when there is a significant difference in the class-conditional distribution between domain features and labels, the domain-invariant features learned by aligning marginal distributions may have limited discriminative ability for unlabeled target instances or even prove counterproductive. To address this issue, we propose a Neighborhood Semantic Aware Learning-based Dynamic Graph Attention Convolution (NSAL-DGAT) approach that learns target semantic information by considering the inter-domain semantic topological structure, thereby improving classifier adaptation for target instances. Specifically, the proposed NSAL framework is designed to capitalize on the insight that after domain feature alignment, some target samples and their neighboring source samples exhibit similar semantics. By leveraging the neighborhood topological structure, we extract and incorporate semantic target features to train a more transferable classifier. Besides, we implement an entropy weighting mechanism to emphasize representative target semantic information, encouraging target instances to prioritize high-confidence individuals within the source neighborhood. We have conducted extensive experiments on the public SEED dataset and our collected the Hearing-Impaired EEG dataset (HIED). The experimental results underscore the efficacy of our proposed NSAL-DGAT approach, showcasing state-of-the-art accuracy in subject-dependent as well as subject-independent scenarios. The source code is available at https://***/YYingDL/NSAL-DGAT. © 2010-2012 IEEE.

关键词： Speech analysis

Collective sleep and activity patterns of college students from wearable devices

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Fudolig, Mikaela Irene Bloomfield, Laura S.P. Price, Matthew Bird, Yoshi M. Hidalgo, Johanna E. Kim, Julia Llorin, Jordan Lovato, Juniper McGinnis, Ellen W. McGinnis, Ryan S. Ricketts, Taylor Stanton, Kathryn Dodds, Peter Sheridan Danforth, Christopher M. Vermont Complex Systems Center MassMutual Center of Excellence for Complex Systems and Data Science University of Vermont BurlingtonVT05405 United States Department of Mathematics and Statistics University of Vermont BurlingtonVT05405 United States Computational Story Lab MassMutual Center of Excellence for Complex Systems and Data Science University of Vermont BurlingtonVT05405 United States Gund Institute for Environment University of Vermont BurlingtonVT05405 United States Department of Psychological Science University of Vermont BurlingtonVT05405 United States Project LEMURS University of Vermont BurlingtonVT05405 United States Department of Electrical and Biomedical Engineering University of Vermont BurlingtonVT05405 United States Wake Forest University School of Medicine Winston-SalemNC27101 United States Rubenstein School of Environment and Natural Resources University of Vermont BurlingtonVT05405 United States Department of Computer Science University of Vermont BurlingtonVT05405 United States Santa Fe Institute New Mexico87501 United States

To optimize interventions for improving wellness, it is essential to understand habits, which wearable devices can measure with greater precision. Using high temporal resolution biometric data taken from the Oura Gen3 ring, we examine daily and weekly sleep and activity patterns of a cohort of young adults (N = 582) in their first semester of college. A high compliance rate is observed for both daily and nightly wear, with slight dips in wear compliance observed shortly after waking up and also in the evening. Most students have a late-night chronotype with a median midpoint of sleep at 5AM, with males and those with mental health impairment having more delayed sleep periods. Social jetlag, or the difference in sleep times between free days and school days, is prevalent in our sample. While sleep periods generally shift earlier on weekdays and later on weekends, sleep duration on both weekdays and weekends is shorter than during prolonged school breaks, suggesting chronic sleep debt when school is in session. Synchronized spikes in activity consistent with class schedules are also observed, suggesting that walking in between classes is a widespread behavior in our sample that substantially contributes to physical activity. Lower active calorie expenditure is associated with weekends and a delayed but longer sleep period the night before, suggesting that for our cohort, active calorie expenditure is affected less by deviations from natural circadian rhythms and more by the timing associated with activities. Our study shows that regular sleep and activity routines may be inferred from consumer wearable devices if high temporal resolution and long data collection periods are available. Copyright © 2024, The Authors. All rights reserved.

关键词： Sleep research

Treatment and Aging Studies of GaAs(111)B Substrates for van der Waals Chalcogenide Film Growth

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Yu, Mingyu Wang, Jiayang Iddawela, Sahani A. McDonough, Molly Thompson, Jessica L. Sinnott, Susan B. Hickey, Danielle Reifsnyder Law, Stephanie Department of Materials Science and Engineering University of Delaware 201 Dupont Hall 127 The Green NewarkDE19716 United States Department of Materials Science and Engineering The Pennsylvania State University University Park PA16802 United States Department of Chemistry The Pennsylvania State University University Park PA16802 United States Materials Research Institute The Pennsylvania State University University Park PA16802 United States Institute for Computational and Data Science The Pennsylvania State University University Park PA16802 United States Penn State Institutes of Energy and the Environment The Pennsylvania State University University Park PA16802 United States 2D Crystal Consortium Material Innovation Platform The Pennsylvania State University University Park PA16802 United States Department of Physics The Pennsylvania State University University Park PA16802 United States

GaAs(111)B is a semiconductor substrate widely used in research andcommercial fields due to its low cost, mature synthesis technology, andexcellent properties for manufacturing electronic devices. It is not only usedto grow three-dimensional (3D) strongly-bonded materials, but has also beenused as a substrate for layered, van der Waals (vdW)-bonded chalcogenide filmgrowth. However, GaAs(111)B wafers cannot be directly used for growingepitaxial vdW chalcogenide films for two reasons: (1) the GaAs surface has asubstantial number of dangling bonds that need to be passivated for vdW layersgrowth;(2) the substrate surface is covered with a thin epi-ready oxide layerwhich must be removed before film growth. In this paper, we optimize the methodfor deoxidizing GaAs(111)B substrates under a Se overpressure and successfullycreate a smooth, deoxidized, and passivated substrate for subsequent growth ofvdW chalcogenide materials. We demonstrate the benefits of this method for thegrowth of vdW chalcogenide thin films using GaSe as a representative of vdWchalcogenides. In addition, we find that severely aged substrates havedifficulty maintaining a smooth surface during the deoxidation and passivationprocess and cause GaSe crystals to nucleate in random shapes and *** describe a method using water droplet testing to determine the age of thesubstrate. Finally, X-ray photoelectron spectroscopy (XPS) characterizationreveals that the natural aging of GaAs(111)B in the air results in an increasein surface oxides, Ga2O3 and As2O3, while exposure to ultraviolet (UV)-ozonenot only enhances the contents of these two oxides but also generates a newoxide, As2O5. Our research contributes to expanding the compatibility ofGaAs(111)B with diverse growth materials and the production of high-qualityheterostructure devices. © 2024, CC BY-NC-ND.

关键词： Substrates

BRIDGE: Benchmarking Large Language Models for Understanding Real-world Clinical Practice Text

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Wu, Jiageng Gu, Bowen Zhou, Ren Xie, Kevin Snyder, Doug Jiang, Yixing Carducci, Valentina Wyss, Richard Desai, Rishi J. Alsentzer, Emily Celi, Leo Anthony Rodman, Adam Schneeweiss, Sebastian Chen, Jonathan H. Romero-Brufau, Santiago Lin, Kueiyu Joshua Yang, Jie Division of Pharmacoepidemiology and Pharmacoeconomics Department of Medicine Brigham and Women’s Hospital Harvard Medical School BostonMA United States Siebel School of Computing and Data Science The Grainger College of Engineering University of Illinois Urbana-Champaign UrbanaIL United States Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology CambridgeMA United States Department of Otorhinolaryngology – Head & Neck Surgery Mayo Clinic RochesterMN United States Department of Biostatistics Harvard T.H. Chan School of Public Health Harvard University BostonMA United States Department of Biomedical Data Science Stanford University Palo AltoCA United States Laboratory for Computational Physiology Massachusetts Institute of Technology CambridgeMA United States Division of Pulmonary Critical Care and Sleep Medicine Beth Israel Deaconess Medical Center BostonMA United States Division of General Internal Medicine Department of Medicine Beth Israel Deaconess Medical Center BostonMA United States Stanford Center for Biomedical Informatics Research Stanford University StanfordCA United States Division of Hospital Medicine Stanford University StanfordCA United States Stanford Clinical Excellence Research Center Stanford University StanfordCA United States Kempner Institute for the Study of Natural and Artificial Intelligence Harvard University MA United States Broad Institute of MIT and Harvard CambridgeMA United States Harvard Data Science Initiative Harvard University CambridgeMA United States

Large language models (LLMs) hold great promise for medical applications and are evolving rapidly, with new models being released at an accelerated pace. However, current evaluations of LLMs in clinical contexts remain limited. Most existing benchmarks rely on medical exam-style questions or PubMed-derived text, failing to capture the complexity of real-world electronic health record (EHR) data. Others focus narrowly on specific application scenarios, limiting their generalizability across broader clinical use. To address this gap, we present BRIDGE, a comprehensive multilingual benchmark comprising 87 tasks sourced from real-world clinical data sources across nine languages. We systematically evaluated 52 state-of-the-art LLMs (including DeepSeek-R1, GPT-4o, Gemini, and Llama 4) under various inference strategies. With a total of 13,572 experiments, our results reveal substantial performance variation across model sizes, languages, natural language processing tasks, and clinical specialties. Notably, we demonstrate that open-source LLMs can achieve performance comparable to proprietary models, while medically fine-tuned LLMs based on older architectures often underperform versus updated general-purpose models. The BRIDGE and its corresponding leaderboard serve as a foundational resource and a unique reference for the development and evaluation of new LLMs in real-world clinical text understanding. © 2025, CC BY-NC-SA.

关键词： Electronic health record

Heterogeneous influence of individuals’ behavior on mask efficacy in gathering environments

学校读者我要写书评

暂无评论

Frontiers of engineering Management 2022年第4期9卷 550-562页

作者： Haochen SUN Xiaofan LIU Zhanwei DU Ye WU Haifeng ZHANG Xiaoke XU College of Information and Communication Engineering Dalian Minzu UniversityDalian 116600China Web Mining Laboratory Department of Media and CommunicationCity University of Hong KongHong KongChina WHO Collaborating Centre for Infectious Disease Epidemiology and Control School of Public HealthLi Ka Shing Faculty of MedicineThe University of Hong KongHong KongChina Laboratory of Data Discovery for Health Hong Kong Science and Technology ParkHong KongChina Computational Communication Research Center Beijing Normal University(Zhuhai)Zhuhai 519087China School of Journalism and Communication Beijing Normal UniversityBeijing 100875China Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education School of Mathematical ScienceAnhui UniversityHefei 230601China

Wearing masks is an easy way to operate and popular measure for preventing *** masks can slow down the spread of viruses,their efficacy in gathering environments involving heterogeneous person-to-person contacts remains ***,we aim to investigate the epidemic prevention effect of masks in different real-life gathering *** study uses four real interpersonal contact datasets to construct four empirical networks to represent four gathering *** transmission of COVID-19 is simulated using the Monte Carlo simulation *** heterogeneity of individuals can cause mask efficacy in a specific gathering environment to be different from the baseline efficacy in general ***,the heterogeneity of gathering environments causes the epidemic prevention effect of masks to *** masks can greatly reduce the probability of clustered epidemics and the infection scale in primary schools,high schools,and ***,the use of masks alone in primary schools and hospitals cannot control *** high schools with social distancing between classes and in workplaces where the interpersonal contact is relatively sparse,masks can meet the need for *** the heterogeneity of individual behavior,if individuals who are more active in terms of interpersonal contact are prioritized for mask-wearing,the epidemic prevention effect of masks can be ***,asymptomatic infection has varying effects on the prevention effect of masks in different *** effect can be weakened or eliminated by increasing the usage rate of masks in high schools and ***,the effect on primary schools and hospitals cannot be *** study contributes to the accurate evaluation of mask efficacy in various gathering environments to provide scientific guidance for epidemic prevention.

关键词： COVID-19 masks behavioral heterogeneity asymptomatic infection

Hyena architecture enables fast and efficient protein language modeling

学校读者我要写书评

暂无评论

iMetaOmics

iMetaOmics 2024年第1期2卷

作者： Yiming Zhang Bian Bian Manabu Okumura Department of Information and Communications Engineering School of Engineering Institute of Science Tokyo Yokohama Japan Department of Computational Biology and Medical Sciences Graduate School of Frontier Sciences The University of Tokyo Kashiwa Japan Artificial Intelligence Research Center National Institute of Advanced Industrial Science and Technology (AIST) Tokyo Japan Department of Data Science School of Frontier Engineering Kitasato University Sagamihara Japan

The emergence of self-supervised deep language models has revolutionized natural language processing tasks and has recently extended its applications to biological sequence analysis. Traditional language models, primarily based on Transformer architectures, demonstrate substantial effectiveness in various applications. However, these models are inherently constrained by the attention mechanism's quadratic computational complexity, , which limits their efficiency and leads to high computational costs. To address these limitations, we introduce ProtHyena, a novel approach that leverages the Hyena operator in protein language modeling. This innovative methodology alternates between subquadratic long convolutions and element-wise gating operations, which circumvents the constraints imposed by attention mechanisms and reduces computational complexity to subquadratic levels. This enables faster and more memory-efficient modeling of protein sequences. ProtHyena can achieve state-of-the-art results and comparable performance in 8 downstream tasks, including protein engineering (protein fluorescence and stability prediction), protein property prediction (neuropeptide cleavage, signal peptide, solubility, disorder, gene function prediction), protein structure prediction, with only 1.6 M parameters. The architecture of ProtHyena represents a highly efficient solution for protein language modeling, offering a promising avenue for fast and efficient analysis of protein sequences. Novel Architecture: ProtHyena applies Hyena architecture for protein language modeling, outperforming traditional attention-based models with lower computational costs and fewer parameters. State-of-the-Art Performance: Achieves leading results across classification and regression tasks, excelling with diverse protein sequences. Efficiency: Optimized for faster training and lower memory usage, enabling scalable modeling of long protein sequences.

关键词： Hyena architecture protein language model protein pre-training