检索结果-内蒙古大学图书馆

arXiv 2020年

作者： Caro, Josue Ortega Ju, Yilong Pyle, Ryan Dey, Sourav Brendel, Wieland Anselmi, Fabio Patel, Ankit B. Department of Neuroscience Baylor College of Medicine HoustonTX77030 United States Quantitative and Computational Bioscience Program Baylor College of Medicine HoustonTX70030 United States Department of Computer Science Rice University HoustonTX77005 United States Manifold AI United States University of Tübingen Germany Department of Electrical and Computer Engineering Rice University HoustonTX70005 United States

Adversarial Attacks are still a significant challenge for neural networks. Recent efforts has shown that adversarial perturbations typically contain high-frequency features, but the root cause of this phenomenon remains unknown. Inspired by theoretical work on linear full-width convolutional models, we hypothesize that convolutional operations with localized kernels are implicitly biased to learn high frequency features, and that this is one of the root causes of high frequency adversarial examples. To test this hypothesis, we analyzed the impact of different choices of linear and nonlinear architectures on the implicit bias of the learned features and the adversarial perturbations, in spatial and frequency domains. We find that, independently of the training dataset, convolutional operations have higher frequency adversarial attacks compared to other architectural parametrizations, and this phenomenon is exacerbated with stronger locality of the convolutional kernels. The explanation for the kernel size dependence involves the Fourier Uncertainty Principle: a spatially-limited (local in the space domain) filter cannot also be frequency-limited (local in the frequency domain). Using larger convolution kernel sizes or avoiding convolutions (e.g. by using Vision Transformers or MLP-style architectures) significantly reduces this high frequency bias. Looking forward, our work strongly suggests that understanding and controlling the implicit bias of architectures will be essential for achieving adversarial robustness. © 2020, CC BY.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Analyzing dominance move (MIP-DoM) indicator for multi- And many-objective optimization

arXiv

引用

arXiv 2020年

作者： do Val Lopes, Claudio Lucio Martins, Flávio Vinícius Cruzeiro Wanner, Elizabeth Fialho Deb, Kalyanmoy Postgraduate Program in Mathematical and Computational Modeling CEFET-MG Brazil Computer Department Centro Federal de Educação Tecnológica de Minas Gerais 30510-000 Brazil Computer Department Centro Federal de Educação Tecnológica de Minas Gerais 30510-000 Brazil Department of Computer Science and Engineering Michigan State University East LansingMI48824 United States

—Dominance move (DoM) is a binary quality indicator that can be used in multi-objective and many-objective optimization to compare two solution sets obtained from different algorithms. The DoM indicator can differentiate the sets for certain important features, such as convergence, spread, uniformity, and cardinality. DoM does not use any reference, and it has an intuitive and physical meaning, similar to the -indicator, and calculates the minimum total move of members of one set so that all elements in another set are to be dominated or identical to at least one member of the first set. Despite the aforementioned properties, DoM is hard to calculate, particularly in higher dimensions. There is an efficient and exact method to calculate it in a bi-objective case only. This work proposes a novel approach to calculate DoM using a mixed integer programming (MIP) approach, which can handle sets with three or more objectives and is shown to overcome the -indicator’s information loss. Experiments, in the bi-objective space, are done to verify the model’s correctness. Furthermore, other experiments, using 3, 5, 10, 15, 20, 25 and 30-objective problems are performed to show how the model behaves in higher-dimensional cases. Algorithms, such as IBEA, MOEA/D, NSGA-III, NSGA-II, and SPEA2 are used to generate the solution sets (however any other algorithms can also be used with the proposed MIP-DoM indicator). Further extensions are discussed to handle certain idiosyncrasies with some solution sets and also to improve the quality indicator and its use for other situations. Copyright © 2020, The Authors. All rights reserved.

关键词： Multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

The bulk van der Waals layered magnet CrSBr is a quasi-1D material

arXiv

引用

arXiv 2022年

作者： Klein, Julian Pingault, Benjamin Florian, Matthias Heißenbüttel, Marie-Christin Steinhoff, Alexander Song, Zhigang Torres, Kierstin Dirnberger, Florian Curtis, Jonathan B. Weile, Mads Penn, Aubrey Deilmann, Thorsten Dana, Rami Bushati, Rezlind Quan, Jiamin Luxa, Jan Sofer, Zdenek Alù, Andrea Menon, Vinod M. Wurstbauer, Ursula Rohlfing, Michael Narang, Prineha Lončar, Marko Ross, Frances M. Department of Materials Science and Engineering Massachusetts Institute of Technology CambridgeMA02139 United States John A. Paulson School of Engineering and Applied Sciences Harvard University CambridgeMA02138 United States QuTech Delft University of Technology Delft2600 GA Netherlands Department of Electrical and Computer Engineering Department of Physics University of Michigan Ann ArborMI48109 United States Institut für Festkörpertheorie Westfälische Wilhelms-Universität Münster Münster48149 Germany Institut für Theoretische Physik Universität Bremen P.O. Box 330 440 Bremen28334 Germany Bremen Center for Computational Materials Science University of Bremen Bremen28359 Germany Department of Physics City College of New York New YorkNY10031 United States College of Letters and Science UCLA Los AngelesCA90095 United States Department of Physics Technical University of Denmark Kgs. LyngbyDK-2800 Denmark MIT.nano Massachusetts Institute of Technology CambridgeMA02139 United States Department of Physics The Graduate Center City University of New York New YorkNY10016 United States Photonics Initiative CUNY Advanced Science Research Center New YorkNY10031 United States Physics Program Graduate Center City University of New York New YorkNY10026 United States Department of Inorganic Chemistry University of Chemistry and Technology Prague Technická 5 Prague 6166 28 Czech Republic Institute of Physics Center for Nanotechnology University of Münster Münster48149 Germany

Correlated quantum phenomena in one-dimensional (1D) systems that exhibit competing electronic and magnetic order are of strong interest for studying fundamental interactions and excitations, such as Tomonaga-Luttinger liquids and topological orders and defects with properties completely different from the quasiparticles expected in their higher-dimensional counterparts. However, clean 1D electronic systems are difficult to realize experimentally, particularly magnetically ordered systems. Here, we show that the van der Waals layered magnetic semiconductor CrSBr behaves like a quasi-1D material embedded in a magnetically ordered environment. The strong 1D electronic character originates from the Cr-S chains and the combination of weak interlayer hybridization and anisotropy in effective mass and dielectric screening with an effective electron mass ratio of meX/meY ∼ 50. This extreme anisotropy experimentally manifests in strong electron-phonon and exciton-phonon interactions, a Peierls-like structural instability and a Fano resonance from a van Hove singularity of similar strength of metallic carbon nanotubes. Moreover, due to the reduced dimensionality and interlayer coupling, CrSBr hosts spectrally narrow (1 meV) excitons of high binding energy and oscillator strength that inherit the 1D character. Overall, CrSBr is best understood as a stack of weakly hybridized monolayers and appears to be an experimentally attractive candidate for the study of exotic exciton and 1D correlated many-body physics in the presence of magnetic order. Copyright © 2022, The Authors. All rights reserved.

关键词： Excitons

来源：评论

学校读者我要写书评

暂无评论

Benchmarking Emergency Department Triage Prediction Models with Machine Learning and Large Public Electronic Health Records

arXiv

引用

arXiv 2021年

作者： Xie, Feng Zhou, Jun Lee, Jin Wee Tan, Mingrui Li, Siqi Rajnthern, Logasan S.O. Chee, Marcel Lucas Chakraborty, Bibhas Wong, An-Kwok Ian Dagan, Alon Ong, Marcus Eng Hock Gao, Fei Liu, Nan Centre for Quantitative Medicine and Programme in Health Services and Systems Research Duke-NUS Medical School Singapore Singapore Singapore Singapore School of Electrical and Electronic Engineering Nanyang Technological University Singapore Singapore Faculty of Medicine Nursing and Health Sciences Monash University VIC Australia Department of Statistics and Data Science National University of Singapore Singapore Singapore Department of Biostatistics and Bioinformatics Duke University DurhamNC United States Division of Pulmonary Allergy and Critical Care Medicine Duke University DurhamNC United States Department of Emergency Medicine Beth Israel Deaconess Medical Center Harvard Medical School BostonMA United States MIT Critical Data Laboratory for Computational Physiology Institute for Medical Engineering and Science Massachusetts Institute of Technology CambridgeMA United States Department of Emergency Medicine Singapore General Hospital Singapore Singapore SingHealth AI Health Program Singapore Health Services Singapore Singapore Institute of Data Science National University of Singapore Singapore Singapore

The demand for emergency department (ED) services is increasing across the globe, particularly during the current COVID-19 pandemic. Clinical triage and risk assessment have become increasingly challenging due to the shortage of medical resources and the strain on hospital infrastructure caused by the pandemic. As a result of the widespread use of electronic health records (EHRs), we now have access to a vast amount of clinical data, which allows us to develop predictive models and decision support systems to address these challenges. To date, however, there are no widely accepted benchmark ED triage prediction models based on large-scale public EHR data. An open-source benchmarking platform would streamline research workflows by eliminating cumbersome data preprocessing, and facilitate comparisons among different studies and methodologies. In this paper, based on the Medical Information Mart for Intensive Care IV Emergency Department (MIMIC-IV-ED) database, we developed a publicly available benchmark suite for ED triage predictive models and created a benchmark dataset that contains over 400,000 ED visits from 2011 to 2019. We introduced three ED-based outcomes (hospitalization, critical outcomes, and 72-hour ED reattendance) and implemented a variety of popular methodologies, ranging from machine learning methods to clinical scoring systems. We evaluated and compared the performance of these methods against benchmark tasks. Our codes are open-source, allowing anyone with MIMIC-IV-ED data access to perform the same steps in data processing, benchmark model building, and experiments. This study provides future researchers with insights, suggestions, and protocols for managing raw data and developing risk triaging tools for emergency care. © 2021, CC BY.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Author Correction: The high-dimensional space of human diseases built from diagnosis records and mapped to genetic loci

引用

Nature computational science 2023年第7期3卷 658页

作者： Gengjie Jia Yu Li Xue Zhong Kanix Wang Milton Pividori Rabab Alomairy Aniello Esposito Hatem Ltaief Chikashi Terao Masato Akiyama Koichi Matsuda David E Keyes Hae Kyung Im Takashi Gojobori Yoichiro Kamatani Michiaki Kubo Nancy J Cox James Evans Xin Gao Andrey Rzhetsky Shenzhen Branch Guangdong Laboratory of Lingnan Modern Agriculture Genome Analysis Laboratory of the Ministry of Agriculture and Rural Affairs Agricultural Genomics Institute at Shenzhen Chinese Academy of Agricultural Sciences Shenzhen China. jiagengjie@***. Computational Bioscience Research Center King Abdullah University of Science and Technology Thuwal Saudi Arabia. Computer Science Program Computer Electrical and Mathematical Sciences and Engineering Division King Abdullah University of Science and Technology Thuwal Saudi Arabia. Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong People's Republic of China. Department of Medicine and Vanderbilt Genetics Institute Vanderbilt University Medical Center Nashville TN US. Department of Medicine Institute of Genomics and Systems Biology Committee on Genomics Genetics and Systems Biology University of Chicago Chicago IL US. Department of Operations Business Analytics and Information Systems University of Cincinnati Cincinnati OH US. Department of Systems Pharmacology and Translational Therapeutics Perelman School of Medicine University of Pennsylvania Philadelphia PA US. Extreme Computing Research Center Computer Electrical and Mathematical Sciences and Engineering Division King Abdullah University of Science and Technology Thuwal Saudi Arabia. College of Computer Science and Engineering University of Jeddah Jeddah Saudi Arabia. HPE HPC/AI EMEA Research Laboratory Bristol UK. RIKEN Center for Integrative Medical Sciences Yokohama Japan. Clinical Research Center Shizuoka General Hospital Shizuoka Japan. Department of Applied Genetics The School of Pharmaceutical Sciences University of Shizuoka Shizuoka Japan. Department of Ophthalmology Graduate School of Medical Sciences Kyushu University Fukuoka Japan. Department of Computational Biology and Medical Sciences Graduate School of Frontier Sciences The University of Tokyo Tokyo Japan. Biological and Environmenta

来源：评论

学校读者我要写书评

暂无评论

Efficient Formulation of Polarizable Gaussian Multipole Electrostatics for Biomolecular Simulations

arXiv

引用

arXiv 2020年

作者： Wei, Haixin Qi, Ruxi Wang, Junmei Cieplak, Piotr Duan, Yong Luo, Ray Departments of Molecular Biology and Biochemistry Chemical and Biomolecular Engineering Materials Science and Engineering and Biomedical Engineering Graduate Program in Chemical and Materials Physics University of California Irvine IrvineCA92697 United States Department of Pharmaceutical Sciences and Computational Chemical Genomics Screening Center School of Pharmacy University of Pittsburgh PittsburghPA15261 United States SBP Medical Discovery Institute 10901 North Torrey Pines Road San diegoCA92037 United States UC Davis Genome Center Department of Biomedical Engineering University of California Davis One Shields Avenue DavisCA95616 United States

Molecular dynamics simulations of biomolecules have been widely adopted in biomedical studies. As classical point-charge models continue to be used in routine biomolecular applications, there have been growing demands on developing polarizable force fields for handling more complicated biomolecular processes. Here we focus on a recently proposed polarizable Gaussian Multipole (pGM) model for biomolecular simulations. A key benefit of pGM is its screening of all short-range electrostatic interactions in a physically consistent manner, which is critical for stable charge-fitting and is needed to reproduce molecular anisotropy. Another advantage of pGM is that each atom’s multipoles are represented by a single Gaussian function or its derivatives, allowing for more efficient electrostatics than other Gaussian-based models. In this study we present an efficient formulation for the pGM model defined with respect to a local frame formed with a set of covalent basis vectors. The covalent basis vectors are chosen to be along each atom’s covalent bonding directions. The new local frame allows molecular flexibility during molecular simulations and facilitates an efficient formulation of analytical electrostatic forces without explicit torque computation. Subsequent numerical tests show that analytical atomic forces agree excellently with numerical finite-difference forces for the tested system. Finally, the new pGM electrostatics algorithm is interfaced with the PME implementation in Amber for molecular simulations under the periodic boundary conditions. To validate the overall pGM/PME electrostatics, we conducted an NVE simulation for a small water box of 512 water molecules. Our results show that, to achieve energy conservation in the polarizable model, it is important to ensure enough accuracy on both PME and induction iteration. It is hoped that the reformulated pGM model will facilitate the development of future force fields based on the pGM electrostatics for applicatio

关键词： Electrostatics

来源：评论

学校读者我要写书评

暂无评论

Skater’s dilemma

arXiv

引用

arXiv 2020年

作者： Ichinose, Genki Miyagawa, Daiki Ito, Junji Masuda, Naoki Department of Mathematical and Systems Engineering Shizuoka University 3-5-1 Johoku Naka-ku Hamamatsu432-8561 Japan Skating Club Yamanashi Gakuin University 2-4-5 Sakaori Kofu Yamanashi400-8575 Japan Department of Mathematics University at Buffalo State University of New York BuffaloNY14260-2900 United States Computational and Data-Enabled Science and Engineering Program University at Buffalo State University of New York BuffaloNY14260-5030 United States

In some athletic races, such as cycling and types of speed skating races, athletes have to complete a relatively long distance at a high speed in the presence of direct opponents. To win such a race, athletes are motivated to hide behind others to suppress energy consumption before a final moment of the race. This situation seems to produce a social dilemma: players want to hide behind others, whereas if a group of players attempts to do so, they may all lose to other players that overtake them. To support that speed skaters are involved in such a social dilemma, we analyzed video footage data for 14 mass start skating races to find that skaters that hid behind others to avoid air resistance for a long time before the final lap tended to win. Furthermore, the finish rank of the skaters in mass start races was independent of the record of the same skaters in time-trial races measured in the absence of direct opponents. The results suggest that how to strategically cope with a skater’s dilemma may be a key determinant for winning long-distance and high-speed races with direct *** Codes 91A80 Copyright © 2020, The Authors. All rights reserved.

关键词： Game theory

来源：评论

学校读者我要写书评

暂无评论

Advancing the Use of Longitudinal Electronic Health Records: Tutorial for Uncovering Real-World Evidence in Chronic Disease Outcomes

引用

Journal of Medical Internet Research 2025年第1期27卷 e71873页

作者： Huang, Feiqing Hou, Jue Zhou, Ningxuan Greco, Kimberly Lin, Chenyu Sweet, Sara Morini Wen, Jun Shen, Lechen Gonzalez, Nicolas Zhang, Sinian Liao, Katherine P. Cai, Tianrun Xia, Zongqi Bourgeois, Florence T. Cai, Tianxi Department of Biostatistics Harvard T.H. Chan School of Public Health Boston MA United States Harvard-MIT Center for Regulatory Science Harvard Medical School Boston MA United States Division of Biostatistics and Health Data Science School of Public Health University of Minnesota Minneapolis MN United States Department of Biomedical Informatics Harvard Medical School Boston MA United States Department of Engineering University of Toronto Toronto ON Canada Division of Rheumatology Inflammation and Immunity Brigham and Women's Hospital Boston MA United States Department of Neurology University of Pittsburgh Pittsburgh PA United States Computational Health Informatics Program (CHIP) Boston Children's Hospital Boston MA United States Department of Pediatrics Harvard Medical School Boston MA United States

Managing chronic diseases requires ongoing monitoring of disease activity and therapeutic responses to optimize treatment plans. With the growing availability of disease-modifying therapies, it is crucial to investigate comparative effectiveness and long-term outcomes beyond those available from randomized clinical trials. We introduce a comprehensive pipeline for generating reproducible and generalizable real-world evidence on disease outcomes by leveraging electronic health record data. The pipeline first generates scalable disease outcomes by linking electronic health record data with registry data containing a small sample of labeled outcomes. It then applies causal analysis using these scalable outcomes to evaluate therapies for chronic diseases. The implementation of the pipeline is illustrated in a case study based on multiple sclerosis. Our approach addresses challenges in real-world evidence generation for disease activity of chronic conditions, specifically the lack of direct observations on key outcomes and biases arising from imperfect or incomplete data. We present advanced machine learning techniques such as semisupervised and ensemble methods to impute missing outcome data, further incorporating steps for calibrated causal analyses and bias correction. ©Feiqing Huang, Jue Hou, Ningxuan Zhou, Kimberly Greco, Chenyu Lin, Sara Morini Sweet, Jun Wen, Lechen Shen, Nicolas Gonzalez, Sinian Zhang, Katherine P Liao, Tianrun Cai, Zongqi Xia, Florence T Bourgeois, Tianxi Cai.

关键词： calibration causal inference chronic disease outcomes data imputation electronic health records longitudinal disease activity machine learning real-world evidence

来源：评论

学校读者我要写书评

暂无评论

Deep neural network for Wannier function centers

arXiv

引用

arXiv 2019年

作者： Zhang, Linfeng Chen, Mohan Wu, Xifan Wang, Han Weinan, E. Car, Roberto Program in Applied and Computational Mathematics Princeton University PrincetonNJ08544 United States CAPT HEDPS College of Engineering Peking University Beijing100871 China Department of Physics Temple University PhiladelphiaPA19122 United States Laboratory of Computational Physics Institute of Applied Physics and Computational Mathematics Huayuan Road 6 Beijing100088 China Department of Mathematics and Program in Applied and Computational Mathematics Princeton University PrincetonNJ08544 United States Beijing Institute of Big Data Research Beijing100871 China Department of Chemistry Department of Physics Program in Applied and Computational Mathematics Princeton Institute for the Science and Technology of Materials Princeton University PrincetonNJ08544 United States

We introduce a deep neural network (DNN) model that assigns the position of the centers of the electronic charge in the snapshots of a molecular dynamics trajectory. The electronic centers are uniquely specified by the unitary transformation that maps the occupied eigenstates onto maximally localized Wannier functions. In combination with deep potential molecular dynamics, a DNN approach to model the atomic potential energy surface at the ab initio density functional level of theory, the scheme makes possible to predict the dielectric properties of insulators for samples and trajectories inaccessible to direct ab initio simulation, without loss of accuracy. As an example, we report calculations of the infrared absorption spectra of light and heavy water at a dispersion inclusive hybrid functional level of theory, finding good agreement with experiment. Extensions to other spectroscopies, like Raman and sum frequency generation, are discussed. Copyright © 2019, The Authors. All rights reserved.

关键词： Quantum chemistry

来源：评论

学校读者我要写书评

暂无评论

Interpretable network propagation with application to expanding the repertoire of human proteins that interact with SARS-CoV-2

arXiv

引用

arXiv 2020年

作者： Law, Jeffrey N. Akers, Kyle Tasnina, Nure Della Santina, Catherine M. Deutsch, Shay Kshirsagar, Meghana Klein-Seetharaman, Judith Crovella, Mark Rajagopalan, Padmavathy Kasif, Simon Murali, T.M. Interdisciplinary Ph.D. Program in Genetics Bioinformatics and Computational Biology BlacksburgVA United States Department of Computer Science Virginia Tech BlacksburgVA United States Department of Biomedical Engineering Boston University BostonMA United States Department of Mathematics University of California Los AngelesCA United States AI for Good Lab Microsoft RedmondWA United States Department of Chemistry Colorado School of Mines GoldenCO United States Department of Computer Science Boston University BostonMA United States Department of Chemical Engineering Virginia Tech BlacksburgVA United States

Background: Network propagation has been widely used for nearly 20 years to predict gene functions and phenotypes. Despite the popularity of this approach, little attention has been paid to the question of provenance tracing in this context, e.g., determining how much any experimental observation in the input contributes to the score of every prediction. Results: We design a network propagation framework with two novel components and apply it to predict human proteins that directly or indirectly interact with SARS-CoV-2 proteins. First, we trace the provenance of each prediction to its experimentally validated sources, which in our case are human proteins experimentally determined to interact with viral proteins. Second, we design a technique that helps to reduce the manual adjustment of parameters by users. We find that for every top-ranking prediction, the highest contribution to its score arises from a direct neighbor in a human protein-protein interaction network. We further analyze these results to develop functional insights on SARS-CoV-2 that expand on known biology such as the connection between endoplasmic reticulum stress, HSPA5, and anti-clotting agents. Conclusions: We examine how our provenance tracing method can be generalized to a broad class of network-based algorithms. We provide a useful resource for the SARS-CoV-2 community that implicates many previously undocumented proteins with putative functional relationships to viral infection. This resource includes potential drugs that can be opportunistically repositioned to target these proteins. We also discuss how our overall framework can be extended to other, newly-emerging viruses. © 2020, CC BY-NC-SA.

关键词： SARS

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：