检索结果-内蒙古大学图书馆

Improving wind speed forecasting at Adama wind farm II in Ethiopia through deep learning algorithms

Case Studies in Chemical and Environmental Engineering 2024年 9卷

作者： Chaka, Mesfin Diro Semie, Addisu Gezahegn Mekonnen, Yedilfana Setarge Geffe, Chernet Amente Kebede, Hailemichael Mersha, Yonas Anose, Fikru Abiko Benti, Natei Ermias Computational Data Science Program College of Natural and Computational Sciences Addis Ababa University P. O. Box 1176 Addis Ababa Ethiopia Center for Environmental Science College of Natural and Computational Sciences Addis Ababa University P. O. Box 1176 Addis Ababa Ethiopia Department of Physics College of Natural and Computational Sciences Addis Ababa University P.O. Box 1176 Addis Ababa Ethiopia Ethiopian Institute of Water Resources (EIWR) Addis Ababa University (AAU) Addis Ababa Ethiopia Department of Physics College of Natural and Computational Sciences Wolkite University Ethiopia

Wind power plays a critical role in supporting Ethiopia's electricity generation, particularly during dry seasons when hydropower availability diminishes. This contribution becomes even more crucial for sustaining industrial growth. However, accurately estimating wind energy remains challenging due to its random and highly variable nature. Deep learning models for time series prediction have been employed to address this issue, and this study evaluates the effectiveness of long short-term memory (LSTM) and gated recurrent unit (GRU) models in predicting wind speed at Adama Wind Farm II in Ethiopia. Model performance was assessed using R², RMSE, and MAE. GRU significantly outperformed LSTM across all metrics and forecasting horizons. For daily forecasting, GRU achieved an R² of 0.9727, exceeding LSTM's 0.9062 by 7.34 %, with corresponding RMSE values of 0.0005 and 0.0018, respectively. Similarly, GRU surpassed LSTM's performance for weekly and monthly forecasting. In weekly forecasting, GRU achieved an R² of 0.9969 compared to LSTM's 0.9930, with RMSE values of 0.0001 for both models. For monthly forecasting, GRU achieved an R² of 0.9330 compared to LSTM's 0.9303, with corresponding RMSE values of 0.0019 for both models. MAE values also followed the same pattern, with GRU consistently demonstrating lower values than LSTM across all forecasting horizons. These findings advance renewable energy and wind power forecasting research, highlighting the adaptability and versatility of LSTM and GRU models for capturing wind behavior. Decision-makers can use these findings to optimize wind power production, enhance grid operation efficiency, and promote sustainable and cost-effective wind energy. LSTM and GRU models have emerged as powerful tools for precise wind speed and wind power forecasting, which are essential for modern wind energy planning systems. They lay the foundation for a more environmentally friendly and sustainable en

关键词： Adama wind farm II Deep learning Ethiopia LSTM and GRU models Renewable energy Wind speed forecasting

来源：评论

学校读者我要写书评

暂无评论

On the particle approximation of lagged Feynman–Kac formulae

引用

Stochastic Processes and their Applications 2025年 188卷

作者： Awadelkarim, Elsiddig Caffarel, Michel Del Moral, Pierre Jasra, Ajay Applied Mathematics and Computational Science Program Computer Electrical and Mathematical Sciences and Engineering Division King Abdullah University of Science and Technology Thuwal23955-6900 Saudi Arabia University of Toulouse CNRS-Lab. de Chimie et Physique Quantiques Toulouse31062 France Centre de Recherche Inria Bordeaux Sud-Ouest Talence33405 France School of Data Science The Chinese University of Hong Kong Shenzhen Shenzhen China

In this paper we examine the numerical approximation of the limiting invariant measure associated with Feynman–Kac formulae. These are expressed in a discrete time formulation and are associated with a Markov chain and a potential function. The typical application considered here is the computation of eigenvalues associated with non-negative operators as found, for example, in physics or particle simulation of rare-events. We focus on a novel lagged approximation of this invariant measure, based upon the introduction of a ratio of time-averaged Feynman–Kac marginals associated with a positive operator iterated l∈N times;a lagged Feynman–Kac formula. This estimator and its approximation using Diffusion Monte Carlo (DMC) are commonly used in the physics literature. In short, DMC is an iterative algorithm involving N∈N particles or walkers simulated in parallel, that undergo sampling and resampling operations. In this work, it is shown that for the DMC approximation of the lagged Feynman–Kac formula, one has an almost sure characterization of the L1-error as the time parameter (iteration) goes to infinity and this is at most of O(exp{−κl}/N), for κ>0. In addition a non-asymptotic in time, and time uniform L1−bound is proved which is O(l/N). We also prove a novel central limit theorem to give a characterization of the exact asymptotic in time variance. This analysis demonstrates that the strategy used in physics, namely, to run DMC with N and l small and, for long time enough, is mathematically justified. Our results also suggest how one should choose N and l in practice. We emphasize that these results are not restricted to physical applications;they have broad relevance to the general problem of particle simulation of the Feynman–Kac formula, which is utilized in a great variety of scientific and engineering fields. © 2025 Elsevier B.V.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

Big data in healthcare:Conceptual network structure,key challenges and opportunities

引用

Digital Communications and Networks 2023年第4期9卷 856-868页

作者： Leonardo B.Furstenaua Pedro Leivas Michele Kremer Sott Michael S.Dohan Jose Ricardo Lopez-Robles Manuel J.Cobo Nicola Luigi Bragazzi Kim-Kwang Raymond Choo Post-Graduate Program in Industrial Engineering Federal University of Rio Grande Do SulPorto Alegre90035-190Brazil University of Santa Cruz Do Sul Av.Independencia 2293Santa Cruz Do SulRSBrazil Business School Unisinos UniversityAv.Dr.Nilo Peçanha 160091330-002Porto AlegreBrazil Faculty of Business Administration Lakehead UniversityThunder BayOntarioCanada Autonomous University of Zacatecas Av.Ramon Lopez Velarde Col.CentroZacatecas98000Mexico Department of Computer Science and Artificial Intelligence Andalusian Research Institute in Data Science and Computational Intelligence(DaSCI)University of GranadaGranada 18071Spain York University ON M3J 1P3Canada Department of Information Systems and Cyber Security University of Texas at San AntonioSan AntonioTX78249USA

Big data is a concept that deals with large or complex data sets by using data analysis tools(e.g.,data mining,machine learning)to analyze information extracted from several sources *** data has attracted wide attention from academia,for example,in supporting patients and health professionals by improving the accuracy of decision-making,diagnosis and disease *** research aimed to perform a Bibliometric Performance and Network Analysis(BPNA)supported by a Scoping Review(SR)to depict the strategic themes,thematic evolution structure,main challenges and opportunities related to the concept of big data applied in the healthcare *** this goal in mind,4857 documents from the Web of science covering the period between 2009 to June 2020 were analyzed with the support of SciMAT *** bibliometric performance showed the number of publications and citations over time,scientific productivity and the geographic distribution of publications and research *** strategic diagram yielded 20 clusters and their relative importance in terms of centrality and *** thematic evolution structure presented the most important themes and how it changes over ***,we presented the main challenges and future opportunities of big data in healthcare.

关键词： Big data Healthcare digitalization Bibliometric Strategic intelligence Co-word analysis SciMAT

来源：评论

学校读者我要写书评

暂无评论

Accuracy of a one-dimensional reduction of dynamical systems on networks

引用

Physical Review E 2022年第2期105卷 024305-024305页

作者： Prosenjit Kundu Hiroshi Kori Naoki Masuda Department of Mathematics State University of New York at Buffalo Buffalo New York 14260-2900 USA Department of Complexity Science and Engineering The University of Tokyo Chiba 277-8561 Japan Computational and Data-Enabled Science and Engineering Program State University of New York at Buffalo Buffalo New York 14260-5030 USA

Resilience is an ability of a system with which the system can adjust its activity to maintain its functionality when it is perturbed. To study resilience of dynamics on networks, Gao et al. [Nature (London) 530, 307 (2016)] proposed a theoretical framework to reduce dynamical systems on networks, which are high dimensional in general, to one-dimensional dynamical systems. The accuracy of this one-dimensional reduction relies on three approximations in addition to the assumption that the network has a negligible degree correlation. In the present study, we analyze the accuracy of the one-dimensional reduction assuming networks without degree correlation. We do so mainly through examining the validity of the individual assumptions underlying the method. Across five dynamical system models, we find that the accuracy of the one-dimensional reduction hinges on the spread of the equilibrium value of the state variable across the nodes in most cases. Specifically, the one-dimensional reduction tends to be accurate when the dispersion of the node's state is small. We also find that the correlation between the node's state and the node's degree, which is common for various dynamical systems on networks, is unrelated to the accuracy of the one-dimensional reduction.

关键词： Complex systems Dynamics of networks Network resilience

来源：评论

学校读者我要写书评

暂无评论

BENCHMARKING COMMUNITY DRUG RESPONSE PREDICTION MODELS: dataSETS, MODELS, TOOLS, AND METRICS FOR CROSS-dataSET GENERALIZATION ANALYSIS

arXiv

引用

arXiv 2025年

作者： Partin, Alexander Vasanthakumari, Priyanka Narykov, Oleksandr Wilke, Andreas Koussa, Natasha Jones, Sara E. Zhu, Yitan Overbeek, Jamie C. Jain, Rajeev Fernando, Gayara Demini Sanchez-Villalobos, Cesar Garcia-Cardona, Cristina Mohd-Yusof, Jamaludin Chia, Nicholas Wozniak, Justin M. Ghosh, Souparno Pal, Ranadip Brettin, Thomas S. Weil, M. Ryan Stevens, Rick L. Division of Data Science and Learning Argonne National Laboratory LemontIL United States Frederick National Laboratory for Cancer Research Cancer Data Science Initiatives Cancer Research Technology Program FrederickMD United States Department of Statistics University of Nebraska–Lincoln LincolnNE United States Department of Electrical & Computer Engineering Texas Tech University LubbockTX United States Division of Computer Computational and Statistical Sciences Los Alamos National Laboratory Los AlamosNM United States Department of Computer Science The University of Chicago ChicagoIL United States

Deep learning (DL) and machine learning (ML) models have shown promise in drug response prediction (DRP), yet their ability to generalize across datasets remains an open question, raising concerns about their real-world applicability. Due to the lack of standardized benchmarking approaches, model evaluations and comparisons often rely on inconsistent datasets and evaluation criteria, making it difficult to assess true predictive capabilities. In this work, we introduce a benchmarking framework for evaluating cross-dataset prediction generalization in DRP models. Our framework incorporates five publicly available drug screening datasets, six standardized DRP models, and a scalable workflow for systematic evaluation. To assess model generalization, we introduce a set of evaluation metrics that quantify both absolute performance (e.g., predictive accuracy across datasets) and relative performance (e.g., performance drop compared to within-dataset results), enabling a more comprehensive assessment of model transferability. Our results reveal substantial performance drops when models are tested on unseen datasets, underscoring the importance of rigorous generalization assessments. While several models demonstrate relatively strong cross-dataset generalization, no single model consistently outperforms across all datasets. Furthermore, we identify CTRPv2 as the most effective source dataset for training, yielding higher generalization scores across target datasets. By sharing this standardized evaluation framework with the community, our study aims to establish a rigorous foundation for model comparison, and accelerate the development of robust DRP models for real-world applications. © 2025, CC BY.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

On the Particle Approximation of Lagged Feynman-Kac Formulae

arXiv

引用

arXiv 2024年

作者： Awadelkarim, Elsiddig Caffarel, Michel Moral, Pierre Del Jasra, Ajay Applied Mathematics and Computational Science Program Computer Electrical and Mathematical Sciences and Engineering Division King Abdullah University of Science and Technology Thuwal23955-6900 Saudi Arabia University of Toulouse CNRS-Lab. de Chimie et Physique Quantiques Toulouse31062 France Centre de Recherche Inria Bordeaux Sud-Ouest Talence33405 France School of Data Science The Chinese University Hong Kong Shenzhen China

In this paper we examine the numerical approximation of the limiting invariant measure associated with Feynman-Kac formulae. These are expressed in a discrete time formulation and are associated with a Markov chain and a potential function. The typical application considered here is the computation of eigenvalues associated with non-negative operators as found, for example, in physics or particle simulation of rare-events. We focus on a novel lagged approximation of this invariant measure, based upon the introduction of a ratio of time-averaged Feynman-Kac marginals associated with a positive operator iterated l ∈ N times;a lagged Feynman-Kac formula. This estimator and its approximation using Diffusion Monte Carlo (DMC) have been extensively employed in the physics literature. In short, DMC is an iterative algorithm involving N ∈ N particles or walkers simulated in parallel, that undergo sampling and resampling operations. In this work, it is shown that for the DMC approximation of the lagged Feynman-Kac formula, one has an almost sure characterization of the L1-error as the time parameter (iteration) goes to infinity and this is at most of O(exp{−κl}/N), for_ κ > 0. In addition a non-asymptotic in time, and time uniform L1−bound is proved which is O(l/√N). We also prove a novel central limit theorem to give a characterization of the exact asymptotic in time variance. This analysis demonstrates that the strategy used in physics, namely, to run DMC with N and l small and, for long time enough, is mathematically justified. Our results also suggest how one should choose N and l in practice. We emphasize that these results are not restricted to physical applications;they have broad relevance to the general problem of particle simulation of the Feynman-Kac formula, which is utilized in a great variety of scientific and engineering fields. © 2024, CC BY.

关键词： Eigenvalues and eigenfunctions

来源：评论

学校读者我要写书评

暂无评论

The Fisher Geometry and Geodesics of the Multivariate Normals, without Differential Geometry

arXiv

引用

arXiv 2023年

作者： Lawson, Brodie A.J. Burrage, Kevin Mengersen, Kerrie Santos, Rodrigo Weber dos Centre for Data Science Queensland University of Technology Australia School of Mathematical Sciences Queensland University of Technology Australia ARC Centre of Excellence for Plant Success in Nature and Agriculture Australia Department of Computer Science University of Oxford United Kingdom Graduate Program in Computational Modeling Universidade Federal de Juiz de Fora Brazil

Choosing the Fisher information as the metric tensor for a Riemannian manifold provides a powerful yet fundamental way to understand statistical distribution families. Distances along this manifold become a compelling measure of statistical distance, and paths of shorter distance improve sampling techniques that leverage a sequence of distributions in their operation. Unfortunately, even for a distribution as generally tractable as the multivariate normal distribution, this information geometry proves unwieldy enough that closed-form solutions for shortest-distance paths or their lengths remain unavailable outside of limited special cases. In this review we present for general statisticians the most practical aspects of the Fisher geometry for this fundamental distribution family. Rather than a differential geometric treatment, we use an intuitive understanding of the covariance-induced curvature of this manifold to unify the special cases with known closed-form solution and review approximate solutions for the general case. We also use the multivariate normal information geometry to better understand the paths or distances commonly used in statistics (annealing, Wasserstein). Given the unavailability of a general solution, we also discuss the methods used for numerically obtaining geodesics in the space of multivariate normals, identifying remaining challenges and suggesting methodological *** Codes 62B11 (primary), 62-02, 62-08 (secondary) © 2023, CC BY.

关键词： Geometry

来源：评论

学校读者我要写书评

暂无评论

Cryptocurrency Analysis using Machine Learning and Deep Learning

Cryptocurrency Analysis using Machine Learning and Deep Lear...

引用

2021 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2021

作者： Thomas, C. Watson, Z. Kim, M. Baidya, A. Lamsal, M. Chowdhury, M.H. Basnet, M. Poudel, K.N. East Central University Department of Mathematics and Computer Science AdaOK United States Middle Tennessee State University Computational Science Program MurfreesboroTN United States University of Memphis Department of Electrical and Computer Engineering MemphisTN United States Embry-Riddle Aeronautical University Department of Mathematics and Data Science FL United States

来源：评论

学校读者我要写书评

暂无评论

ζ-QVAE: A Quantum Variational Autoencoder utilizing Regularized Mixed-state Latent Representations

arXiv

引用

arXiv 2024年

作者： Wang, Gaoyuan Warrell, Jonathan Emani, Prashant S. Gerstein, Mark Program in Computational Biology and Bioinformatics Yale University New HavenCT06520 United States Department of Molecular Biophysics and Biochemistry Yale University New HavenCT06520 United States Department of Computer Science Yale University New HavenCT06520 United States Department of Statistics & Data Science Yale University New HavenCT06520 United States Department of Biomedical Informatics & Data Science Yale University New HavenCT06520 United States NEC Laboratories America PrincetonNJ08540 United States

A major challenge in near-term quantum computing is its application to large real-world datasets due to scarce quantum hardware resources. One approach to enabling tractable quantum models for such datasets involves compressing the original data to manageable dimensions while still representing essential information for downstream analysis. In classical machine learning, variational autoencoders (VAEs) facilitate efficient data compression, representation learning for subsequent tasks, and novel data generation. However, no model has been proposed that exactly captures all of these features for direct application to quantum data on quantum computers. Some existing quantum models for data compression lack regularization of latent representations, thus preventing direct use for generation and control of generalization. Others are hybrid models with only some internal quantum components, impeding direct training on quantum data. To bridge this gap, we present a fully quantum framework, ζ-QVAE, which encompasses all the capabilities of classical VAEs and can be directly applied for both classical and quantum data compression. Our model utilizes regularized mixed states to attain optimal latent representations. It accommodates various divergences for reconstruction and regularization. Furthermore, by accommodating mixed states at every stage, it can utilize the full-data density matrix and allow for a"global"training objective. Doing so, in turn, makes efficient optimization possible and has potential implications for private and federated learning. In addition to exploring the theoretical properties of ζ-QVAE, we demonstrate its performance on representative genomics and synthetic data. Our results consistently indicate that ζ-QVAE exhibits similar or better performance compared to matched classical models. © 2024, CC BY-NC-ND.

关键词： Large datasets

来源：评论

学校读者我要写书评

暂无评论

Quantifying gender imbalance in East Asian academia: Research career and citation practice

arXiv

引用

arXiv 2023年

作者： Nakajima, Kazuki Liu, Ruodan Shudo, Kazuyuki Masuda, Naoki Department of Mathematical and Computing Science Tokyo Institute of Technology Meguro-ku Tokyo152-8552 Japan Department of Mathematics State University of New York at Buffalo Buffalo14260 United States Academic Center for Computing and Media Studies Kyoto University Sakyo-ku Kyoto606-8501 Japan Computational and Data-Enabled Science and Engineering Program State University of New York at Buffalo Buffalo14260 United States Center for Computational Social Science Kobe University Kobe657-8501 Japan

Gender imbalance in academia has been confirmed in terms of a variety of indicators, and its magnitude often varies from country to country. Europe and North America, which cover a large fraction of research workforce in the world, have been the main geographical regions for research on gender imbalance in academia. However, the academia in East Asia, which accounts for a substantial fraction of research, may be exposed to strong gender imbalance because Asia has been facing persistent and stronger gender imbalance in society at large than Europe and North America. Here we use publication data between 1950 and 2020 to analyze gender imbalance in academia in China, Japan, and South Korea in terms of the number of researchers, their career, and citation practice. We found that, compared to the average of the other countries, gender imbalance is larger in these three East Asian countries in terms of the number of researchers and their citation practice and additionally in Japan in terms of research career. Moreover, we found that Japan has been exposed to the larger gender imbalance than China and South Korea in terms of research career and citation practice. Copyright © 2023, The Authors. All rights reserved.

关键词： Geographical regions

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

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

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：