检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Merleau, Nono S.C. O’Malley, Miguel Roldán, Érika Mukherjee, Sayan Center for Scalable Data Analytics and Artificial Intelligence Department of Computer Science Universität Leipzig Germany Max Planck Institute for Mathematics in the Sciences Germany Duke University Departments of Statistical Science Mathematics Computer Science and Biostatistics & Bioinformatics United States

Higher-dimensional sliding puzzles are constructed on the vertices of a d-dimensional hypercube, where 2d−l vertices are distinctly coloured. Rings with the same colours are initially set randomly on the vertices of the hypercube. The goal of the puzzle is to move each of the 2d − l rings to pre-defined target vertices on the cube. In this setting, the k-rule constraint represents a generalisation of edge collision for the movement of colours between vertices, allowing movement only when a hypercube face of dimension k containing a ring is completely free of other rings. Starting from an initial configuration, what is the minimum number of moves needed to make ring colours match the vertex colours? An algorithm that provides us with such a number is called God’s algorithm. When such an algorithm exists, it does not have a polynomial time complexity, at least in the case of the 15-puzzle corresponding to k = 1 in the cubical puzzle. This paper presents a comprehensive computational study of different scenarios of the higher-dimensional puzzle. A benchmark of three computational techniques, an exact algorithm (the A* search) and two approximately optimal search techniques (an evolutionary algorithm (EA) and reinforcement learning (RL)) is presented in this work. The experiments show that all three methods can successfully solve the puzzle of dimension three for different face dimensions and across various difficulty levels. When the dimension increases, the A* search fails, and RL and EA methods can still provide a generally acceptable solution, i.e. a distribution of a number of moves with a median value of less than 30. Overall, the EA method consistently requires less computational time, while failing in most cases to minimise the number of moves for the puzzle dimensions d = 4 and d = 5. © 2024, CC BY.

关键词： Evolutionary algorithms

来源：评论

学校读者我要写书评

暂无评论

Ensemble Machine Learning Methods for better Dynamic Assessment of Transformer Status

引用

Journal of The Institution of Engineers (India): Series B 2021年第5期102卷 1113-1122页

作者： Ghosh, Soham Dutta, Sreejata Department of Electrical Engineering Arizona State University TempeAZ85281 United States Department of Biostatistics & Data Science University of Kansas Medical Center Kansas CityKS66160 United States

Analyzing dissolved gases in the transformer's mineral oil helps to detect and classify the systemic faults in electric power transformers. Formerly, empirical methods such as Rogers ratio, Duval triangles 1–4–5, and pentagons 1–2 were used for transformer fault classification. Loose fit for every transformer type is one of the most prominent disadvantages of conventional methods. Formulating robust machine learning algorithms, such as the decision trees, can significantly overcome the loose fit issue. This paper focuses on implementing four different decision tree algorithms, including a regular decision tree classifier, a bagging classifier, a boosting classifier, and a stacking classifier to classify six different transformer fault types distinctly. Further, this study shows that the efficacy and accuracy of the four mentioned classifiers could be far exceeded when combined using a wisdom of the crowd approach. The wisdom of the crowd approach essentially merges the predicted classes from the four individual classifiers and decides on the final prediction via a hard-voting routine. The computational evaluation revealed that the given voting approach could significantly improve power transformers' online diagnostic accuracy up to 91%, thus aiding early forecast of power transformers' preventive maintenance. © 2021, The Institution of Engineers (India).

关键词：

来源：评论

学校读者我要写书评

暂无评论

UMVUE-DR: Uniformly Minimum Variance Unbiased Doubly Robust Learning with Reduced Bias and Variance for Debiased Recommendation

UMVUE-DR: Uniformly Minimum Variance Unbiased Doubly Robust ...

引用

IEEE International Conference on data Mining Workshops (ICDM Workshops)

作者： Chunyuan Zheng Taojun Hu Chuan Zhou Xiao-Hua Zhou School of Mathematical Sciences Peking University Beijing China Department of Biostatistics Peking University Beijing China Center for Data Science Peking University Beijing China Beijing International Center for Mathematical Research Peking University Beijing China

ISBN: (数字)9798331530631

ISBN: (纸本)9798331530648

Recommender systems (RS) are essential for personalized experiences across e-commerce, social media, and various online platforms. due to users are free to choose items to rate, there is a distributional shift between the observed ratings and the ratings among all users and all items, which is also known as the selection bias. Doubly robust (DR) methods have been proposed to address this issue with some promising success. In recent years, many methods have been proposed that aim to further control the bias and variance of the DR approach when learning the propensity model. However, we find that these methods are unable to achieve unbiased estimation of bias and variance and have large estimation errors at either small or large propensity scores. To bridge this gap, this paper proposes a uniformly minimum variance unbiased DR learning method with reduced bias and variance. Experiments on three real-world datasets show the effectiveness of our method.

关键词： Learning systems Bridges Estimation error Social networking (online) Conferences Computational efficiency Electronic commerce data mining Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Effect of tracheostomy timing on outcomes in patients with traumatic brain injury

引用

Baylor University Medical center Proceedings 2022年第5期35卷 621-628页

作者： Mubashir, Talha Lai, Hongyin Oduguwa, Emmanuella Chaudhry, Rabail Balogh, Julius Williams, George W. Maroufy, Vahed Department of Anesthesiology McGovern Medical School University of Texas Health Science Center at Houston (UT Health) Houston TX United States Department of Biostatistics and Data Science School of Public Health University of Texas Health Science Center at Houston (UT Health) Houston TX United States Department of Anesthesiology University of Toronto University Health Network Toronto ON Canada Department of Anesthesiology and Critical Care University of Arkansas Medical Center Little Rock AR United States

Tracheostomy following severe traumatic brain injury (TBI) is common, yet the outcomes associated with tracheostomy timing are unclear. The objective of this study was to assess hospital outcomes of tracheostomy timing in TBI patients. We retrospectively analyzed data from the National Inpatient Sample database of adult patients aged ≥18 years with a primary diagnosis of TBI. Indexed hospitalizations of TBI patients who underwent either percutaneous or surgical tracheostomy between 1995 and 2015 in the United States were included. The interventional groups were 1) early tracheostomy (≤7 days) vs standard tracheostomy (8–14 days), vs late tracheostomy (≥15 days), and 2) tracheostomy vs no tracheostomy. Propensity score matching and conditional logistic regression models were used to analyze in-hospital mortality, length of hospitalization, and in-hospital complications among TBI patients in relation to tracheostomy timing. The risk of in-hospital mortality was 35% lower in patients who underwent tracheostomy vs those who did not (odds ratio 0.65;P < 0.001). Patients who underwent early tracheostomy had a higher risk of in-hospital mortality compared to standard tracheostomy (odds ratio 1.69;P < 0.001) or late tracheostomy (odds ratio 1.80;P < 0.001). An early tracheostomy was associated with a shorter mean hospital length of stay (27 days) compared to standard (36 days) or late tracheostomy (48 days). © Copyright © 2022 Baylor University Medical center.

关键词： adults early tracheostomy late tracheostomy traumatic brain injury

来源：评论

学校读者我要写书评

暂无评论

Clmr: An Advanced Attention Convolutional Network with Custom Activation Functions for Speech Emotion Recognition

SSRN

引用

SSRN 2024年

作者： Zhao, Puyang Zhang, Huiyun Yue, Zhiyi Li, Zongjin Department of Biostatistics and Data Science University of Texas Health Science Center HoustonTX77030 United States The School of Software Henan University Kaifeng475004 China College of Science North China University of Science and Technology Tangshan063210 China Qinghai Normal University China

Speech Emotion Recognition (SER) is pivotal in artificial intelligence and human-computer interaction, particularly within public health applications. We propose a novel Convolutional Long Short-Term Memory Multi-Head Attention Residual (CLMR) network to enhance SER performance. The CLMR model integrates a custom activation function, Custom GELU, combining GELU and LeakyReLU to improve nonlinear representation capabilities. Incorporating LSTM networks, multi-head attention mechanisms, and residual connections allows the model to effectively capture temporal and frequency features in speech signals. Additionally, a comprehensive feature combination strategy integrating traditional and nonlinear dynamic features is adopted to improve emotion classification accuracy. Experiments on the BodEMODB, EMODB, SAVEE, and CASIA datasets demonstrate that the CLMR model outperforms existing methods across key evaluation metrics, highlighting its potential in public health applications. This study contributes a novel framework for SER, advancing affective computing, enhancing human-computer interaction, and offering significant benefits for public health applications. © 2024, The Authors. All rights reserved.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Surveillance of Noncommunicable Diseases:Opportunities in the Era of Big data

引用

Health data science 2022年第1期2022卷 153-155页

作者： Pengfei Li Lin Ma Jue Liu Luxia Zhang Advanced Institute of Information Technology Peking UniversityHangzhou 311215China National Institute of Health Data Science Peking UniversityBeijing 100191China Peking University Health Science Center Beijing 100191China Department of Epidemiology and Biostatistics School of Public HealthPeking UniversityBeijing 100191China

*** Noncommunicable diseases(NCDs)are the leading cause of mortality,accounting for 70%of deaths worldwide,and have become one of the major challenges for human sustainable development in the 21st century[1].The pandemic of coronavirus disease 2019(COVID-19)interrupted NCD services in 75%of the countries—public health campaigns and NCD surveillance efforts gave way to pandemic control[2],which further imposes challenges to fight *** the key components of strategy for the prevention and control of NCDs,surveillance is a crucial one to track and monitor the major risk factors and intervention effects,which is defined as“the ongoing systematic collection,analysis,and interpretation of health data essential to the planning,implementation,and evaluation of public health practice,closely integrated with timely dissemination of these data to those who need to know”[3,4].The capacity to undertake surveillance varied substantially and was inadequate in many low-income and middle-income *** is hoped that integration of multiple existing data sources into a comprehensive NCD surveillance framework could enhance the NCD surveillance efforts[5].

关键词： income prevention surveillance

来源：评论

学校读者我要写书评

暂无评论

Comparison of Effect Size Measures for Mediation Analysis of Survival Outcomes with Application to the Framingham Heart Study

arXiv

引用

arXiv 2022年

作者： Shi, Bin Huang, Xuelin Wei, Peng Department of Biostatistics The University of Texas MD Anderson Cancer Center HoustonTX United States Department of Biostatistics and Data Science The University of Texas Health Science Center at Houston HoustonTX United States

There is an increasing trend of research in mediation analysis for survival outcomes. Such analyses help researchers to better understand how exposure affects disease outcomes through mediators. However, due to censored observations in survival outcomes, it is not straightforward to extend mediation analysis from linear models to survival outcomes. In this article, we extend a mediation effect size measure based on R2 in linear regression to survival outcomes. Due to multiple definitions of R2 for survival models, we compare and evaluate five R2 measures for mediation analysis. Based on extensive simulations, we recommend two R2 measures with good operating characteristics. We illustrate the utility of the R2-based mediation measures by analyzing the mediation effects of multiple lifestyle risk factors on the relationship between environmental exposures and time to coronary heart disease and all-cause mortality in the Framingham Heart Study. © 2022, CC BY-NC-ND.

关键词： Heart

来源：评论

学校读者我要写书评

暂无评论

Recent advances in deep learning and language models for studying the microbiome

arXiv

引用

arXiv 2024年

作者： Yan, Binhao Nam, Yunbi Li, Lingyao Deek, Rebecca A. Li, Hongzhe Ma, Siyuan Department of Biostatistics Epidemiology and Informatics Perelman School of Medicine University of Pennsylvania PhiladelphiaPA United States Department of Biostatistics Vanderbilt University Medical Center NashvilleTN United States School of Information University of South Florida TampaFL United States Department of Biostatistics and Health Data Science University of Pittsburgh PittsburghPA United States

Recent advancements in deep learning, particularly large language models (LLMs), made significant impact on how researchers study microbiome and metagenomics data. Microbial protein and genomic sequences, like natural languages, form a language of life, enabling the adoption of LLMs to extract useful insights from complex microbial ecologies. In this paper, we review applications of deep learning and language models in analyzing microbiome and metagenomics data. We focus on problem formulations, necessary datasets, and the integration of language modeling techniques. We provide an extensive overview of protein/genomic language modeling and their contributions to microbiome studies. We also discuss applications such as novel viromics language modeling, biosynthetic gene cluster prediction, and knowledge integration for metagenomics studies. © 2024, CC BY.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Covariate Adjustment in Randomized Experiments Motivated by Higher-Order Influence Functions

arXiv

引用

arXiv 2024年

作者： Zhao, Sihui Wang, Xinbo Liu, Lin Zhang, Xin School of Mathematical Sciences CMA-Shanghai Institute of Natural Sciences MOE-LSC Shanghai Jiao Tong University China Department of Bioinformatics and Biostatistics School of Life Sciences Shanghai Jiao Tong University China SJTU-Yale Joint Center for Biostatistics and Data Science Shanghai Jiao Tong University China Global Biometrics & Data Management Pfizer Inc

Higher-Order Influence Functions (HOIF), developed in a series of papers over the past twenty years, is a fundamental theoretical device for constructing rate-optimal causal-effect estimators from observational studies. However, the value of HOIF for analyzing well-conducted randomized controlled trials (RCTs) has not been explicitly explored. In the recent U.S. Food and Drug Administration (FDA) and European Medicines Agency (EMA) guidelines on the practice of covariate adjustment in analyzing RCTs, in addition to the simple, unadjusted difference-in-mean estimator, it was also recommended to report the estimator adjusting for baseline covariates via a simple parametric working model, such as a linear model. In this paper, we show that a HOIF-motivated estimator for the treatment-specific mean has significantly improved statistical properties compared to popular adjusted estimators in practice when the number of baseline covariates p is relatively large compared to the sample size n. We also characterize the conditions under which the HOIF-motivated estimator improves upon the unadjusted one. Furthermore, we demonstrate that a novel debiased adjusted estimator proposed recently by Lu et al. is, in fact, another HOIF-motivated estimator in disguise. Numerical and empirical studies are conducted to corroborate our theoretical findings. Copyright © 2024, The Authors. All rights reserved.

关键词： Controlled drug delivery

来源：评论

学校读者我要写书评

暂无评论

DenseFormer-MoE: A Dense Transformer Foundation Model with Mixture of Experts for Multi-Task Brain Image Analysis

引用

IEEE Transactions on Medical Imaging 2025年 PP卷 PP页

作者： Ding, Rizhi Lu, Hui Liu, Manhua Shanghai Jiao Tong University MoE Laboratory of ArtificialIntelligence Al Institute School of Electronic Information and Electrical Engineering Shanghai200240 China Shanghai Jiao Tong University National Center for Translational Medicine SJTU-Yale Joint Center of Biostatistics and Data Science 800 Dongchuan Rd Shanghai200240 China

Deep learning models have been widely investigated for computing and analyzing brain images across various downstream tasks such as disease diagnosis and age regression. Most existing models are tailored for specific tasks and diseases, posing a challenge in developing a foundation model for diverse tasks. This paper proposes a Dense Transformer Foundation Model with Mixture of Experts (DenseFormer-MoE), which integrates dense convolutional network, Vision Transformer and Mixture of Experts (MoE) to progressively learn and consolidate local and global features from T1-weighted magnetic resonance images (sMRI) for multiple tasks including diagnosing multiple brain diseases and predicting brain age. First, a foundation model is built by combining the vision Transformer with Densenet, which are pre-trained with Masked Autoencoder and self-supervised learning to enhance the generalization of feature representations. Then, to mitigate optimization conflicts in multi-task learning, MoE is designed to dynamically select the most appropriate experts for each task. Finally, our method is evaluated on multiple renowned brain imaging datasets including UK Biobank (UKB), Alzheimer’s Disease Neuroimaging Initiative (ADNI), and Parkinson’s Progression Markers Initiative (PPMI). Experimental results and comparison demonstrate that our method achieves promising performances for prediction of brain age and diagnosis of brain diseases. © 2025 IEEE. All rights reserved.

关键词： Self-supervised learning

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：