检索结果-内蒙古大学图书馆

35th Conference on Learning Theory (COLT)

作者： Bartlett, Peter Indyk, Piotr Wagner, Tal Univ Calif Berkeley Berkeley CA 94720 USA MIT Cambridge MA 02139 USA Microsoft Res Redmond WA USA

data-driven algorithms can adapt their internal structure or parameters to inputs from unknown application-specific distributions, by learning from a training sample of inputs. Several recent works have applied this approach to problems in numerical linear algebra, obtaining significant empirical gains in performance. However, no theoretical explanation for their success was known. In this work we prove generalization bounds for those algorithms, within the PAC-learning framework for data-driven algorithm selection proposed by Gupta and Roughgarden (SICOMP 2017). Our main results are closely matching upper and lower bounds on the fat shattering dimension of the learning-based low rank approximation algorithm of Indyk et al. (NeurIPS 2019). Our techniques are general, and provide generalization bounds for many other recently proposed data-driven algorithms in numerical linear algebra, covering both sketching-based and multigrid-based methods. This considerably broadens the class of data-driven algorithms for which a PAC-learning analysis is available.

关键词： data-driven algorithms low rank approximation numerical linear algebra PAC-learning pseudo-dimension fat shattering dimension sketching multigrid

来源：评论

学校读者我要写书评

暂无评论

Exposures to select risk factors can be estimated from a continuous stream of inertial sensor measurements during a variety of lifting-lowering tasks

引用

ERGONOMICS 2024年第11期67卷 1596-1611页

作者： Lim, Sol Virginia Tech Dept Ind & Syst Engn Blacksburg VA 24061 USA

Wearable inertial measurement units (IMUs) are used increasingly to estimate biomechanical exposures in lifting-lowering tasks. The objective of the study was to develop and evaluate predictive models for estimating relative hand loads and two other critical biomechanical exposures to gain a comprehensive understanding of work-related musculoskeletal disorders in lifting. We collected 12,480 lifting-lowering phases from 26 subjects (15 men and 11 women) performing manual lifting-lowering tasks with hand loads (0-22.7 kg) at varied workstation heights and handling modes. We implemented a Hierarchical model, that sequentially classified risk factors, including workstation height, handling mode, and relative hand load. Our algorithm detected lifting-lowering phases (>97.8%) with mean onset errors of 0.12 and 0.2 seconds for lifting and lowering phases. It estimated workstation height (>98.5%), handling mode (>87.1%), and relative hand load (mean absolute errors of 5.6-5.8%) across conditions, highlighting the benefits of data-driven models in deriving lifting-lowering occurrences, timing, and critical risk factors from continuous IMU-based kinematics. Practitioner summary: The study developed and validated algorithms for detecting and predicting exposure to various risk factors during diverse lifting-lowering tasks. These factors encompass the occurrence, timing, workstation height, handling mode, and relative hand position. This approach facilitates the extraction of contextual information related to lifting tasks conducted in real-world settings through a continuous stream of inertial sensor measurements. Consequently, it can enable automated risk assessment for lifting activities in the field.

关键词： Inertial sensor measurements lifting-lowering risk factors estimation biomechanical exposures data-driven algorithms

来源：评论

学校读者我要写书评

暂无评论

A PV Power Forecasting Based on Mechanism Model-driven and Stacking Model Fusion

引用

JOURNAL OF ELECTRICAL ENGINEERING & TECHNOLOGY 2024年第8期19卷 4683-4697页

作者： Chen, Fan Ding, Jinjin Zhang, Qian Wu, Junjie Lei, Fan Liu, Yifan State Grid Anhui Elect Power Co Ltd Hefei Peoples R China State Grid Anhui Elect Power Co Ltd Elect Power Res Inst Hefei Peoples R China Anhui Univ Collaborat Innovat Ctr Ind Energy Saving & Power Q Hefei Peoples R China Anhui Univ Sch Elect Engn & Automat Hefei Peoples R China

Accurate short-term forecasting of photovoltaic power generation is crucial for power dispatching, capacity analysis, and unit commitment. Existing data-driven prediction algorithms have a certain impact on calculation speed and prediction accuracy, but they fail to consider the internal mechanism of photovoltaic power generation and have the risk of generalization. First, the fuzzy C-means clustering (FCM) algorithm method was used for preprocessing of the PV sample set. The sample points with variability were categorized into different sample sets with less variability. Second, the photovoltaic mechanism model is added to the first layer learner of the Stacking framework to form a one-layer learner of the Long Short-Term Memory (LSTM) neural network, Light Gradient Boosting model (LGBM), and mechanism-driven model. The mechanistic model limits PV generation to a reasonable range as a prediction constraint for the data-driven model. The proposed model can seize the useful inherent information from the mechanism model and utilize the ability of data analysis to extract the inexplicit linear relationship. Finally, the PV power and weather observation data collected from photovoltaic power stations located in a certain place in Germany are used to verify the effectiveness of the proposed method.

关键词： Weather classification Photovoltaic power forecast Mechanism model data-driven algorithms Stacking model fusion

来源：评论

学校读者我要写书评

暂无评论

Variable Dynamic Mode Decomposition For Estimating Time Eigenvalues in Nuclear Systems

引用

NUCLEAR SCIENCE AND ENGINEERING 2023年第8期197卷 1769-1778页

作者： Smith, Ethan Variansyah, Ilham McClarren, Ryan Univ Notre Dame Dept Aerosp & Mech Engn Fitzpatrick Hall Notre Dame IN 46556 USA

We present a new approach to calculating time eigenvalues of the neutron transport operator (also known as $$\alpha $$alpha eigenvalues) by extending the dynamic mode decomposition (DMD) to allow for nonuniform time steps. The new method, called variable dynamic mode decomposition (VDMD), is shown to be accurate when computing eigenvalues for systems that were infeasible with DMD due to a large separation in timescales (such as those that occur in delayed supercritical systems). The $$\alpha $$alpha eigenvalues of an infinite medium neutron transport problem with delayed neutrons, and consequently having multiple, very different relevant timescales, are computed. Furthermore, VDMD is shown to be of similar accuracy to the original DMD approach when computing eigenvalues in other systems where the previously studied DMD approach can be used.

关键词： Time eigenvalues data-driven algorithms time-dependent neutronics

来源：评论

学校读者我要写书评

暂无评论

Massively Digitized Power Grid: Opportunities and Challenges of Use-Inspired AI

引用

PROCEEDINGS OF THE IEEE 2023年第7期111卷 762-787页

作者： Xie, Le Zheng, Xiangtian Sun, Yannan Huang, Tong Bruton, Tony Texas A&M Univ Dept Elect & Comp Engn College Stn TX 77843 USA Oncor Elect Delivery Dallas TX 75202 USA MIT Lab Informat & Decis Syst 77 Massachusetts Ave Cambridge MA 02139 USA

This article presents a use-inspired perspective of the opportunities and challenges in a massively digitized power grid. It argues that the intricate interplay of data availability, computing capability, and artificial intelligence (AI) algorithm development are the three key factors driving the adoption of digitized solutions in the power grid. The impact of these three factors on critical functions of power system operation and planning practices is reviewed and illustrated with industrial practice case studies. Open challenges and research opportunities for data, computing, and AI algorithms are articulated within the context of the power industry's tremendous decarbonization efforts.

关键词： Artificial intelligence Power grids Security Power systems Low-carbon economy Renewable energy sources Reliability Artificial intelligence (AI) data-driven algorithms decarbonization industry use cases machine learning power grid

来源：评论

学校读者我要写书评

暂无评论

On Matching, and Even Rectifying, Dynamical Systems through Koopman Operator Eigenfunctions

引用

SIAM JOURNAL ON APPLIED DYNAMICAL SYSTEMS 2018年第2期17卷 1925-1960页

作者： Bollt, Erik M. Li, Qianxiao Dietrich, Felix Kevrekidis, Ioannis Clarkson Univ Dept Math Dept Elect & Comp Engn Dept Phys Potsdam NY 13699 USA Agcy Sci Technol & Res Inst High Performance Comp Singapore 138632 Singapore Johns Hopkins Univ Dept Chem & Biomol Engn Dept Appl Math & Stat Baltimore MD 21218 USA JHMI Baltimore MD 21218 USA

Matching dynamical systems, through different forms of conjugacies and equivalences, has long been a fundamental concept, and a powerful tool, in the study and classification of nonlinear dynamic behavior (e.g., through normal forms). In this paper we will argue that the use of the Koopman operator and its spectrum is particularly well suited for this endeavor, both in theory, but also especially in view of recent data-driven algorithm developments. We believe, and document through illustrative examples, that this can nontrivially extend the use and applicability of the Koopman spectral theoretical and computational machinery beyond modeling and prediction, towards what can be considered as a systematic discovery of "Cole-Hopf-type" transformations for dynamics.

关键词： Koopman operator rectification conjugacy flow box DMD EDMD dynamical systems data-driven algorithms

来源：评论

学校读者我要写书评

暂无评论

A machine-learning minimal-residual (ML-MRes) framework for goal-oriented finite element discretizations

引用

COMPUTERS & MATHEMATICS WITH APPLICATIONS 2021年 95卷 186-199页

作者： Brevis, Ignacio Muga, Ignacio van der Zee, Kristoffer G. Pontificia Univ Catolica Valparaiso Inst Matemat Valparaiso Chile Univ Nottingham Sch Math Sci Nottingham England

We introduce the concept of machine-learning minimal-residual (ML-MRes) finite element discretizations of partial differential equations (PDEs), which resolve quantities of interest with striking accuracy, regardless of the underlying mesh size. The methods are obtained within a machine-learning framework during which the parameters defining the method are tuned against available training data. In particular, we use a provably stable parametric Petrov-Galerkin method that is equivalent to a minimal-residual formulation using a weighted norm. While the trial space is a standard finite element space, the test space has parameters that are tuned in an off-line stage. Finding the optimal test space therefore amounts to obtaining a goal-oriented discretization that is completely tailored towards the quantity of interest. We use an artificial neural network to define the parametric family of test spaces. Using numerical examples for the Laplacian and advection equation in one and two dimensions, we demonstrate that the ML-MRes finite element method has superior approximation of quantities of interest even on very coarse meshes. (C) 2020 Elsevier Ltd. All rights reserved.

关键词： Goal-oriented finite elements Machine-learning acceleration Residual minimization Petrov-Galerkin method Weighted inner-products data-driven algorithms

来源：评论

学校读者我要写书评

暂无评论

Optimal Pricing with a Single Point

引用

MANAGEMENT SCIENCE 2023年第10期69卷 5866-5882页

作者： Allouah, Amine Bahamou, Achraf Besbes, Omar Columbia Univ Dept Ind Engn & Operat Res New York NY 10027 USA Columbia Univ Grad Sch Business Decis Risk & Operat Div New York NY 10027 USA

Historical data are typically limited. We study the following fundamental data-driven pricing problem. How can/should a decision maker price its product based on data at a single historical price? How valuable is such data? We consider a decision maker who optimizes over (potentially randomized) pricing policies to maximize the worst-case ratio of the garnered revenue compared to an oracle with full knowledge of the distribution of values, when the latter is only assumed to belong to a broad nonparametric set. In particular, our framework applies to the widely used regular and monotone nondecreasing hazard rate (mhr) classes of distributions. For settings where the seller knows the exact probability of sale associated with one historical price or only a confidence interval for it, we fully characterize optimal performance and near-optimal pricing algorithms that adjust to the information at hand. The framework we develop is general and allows to characterize optimal performance for deterministic or more general randomized mechanisms and leads to fundamental novel insights on the value of data for pricing. As examples, against mhr distributions, we show that it is possible to guarantee 85% of oracle performance if one knows that half of the customers have bought at the historical price, and if only 1% of the customers bought, it still possible to guarantee 51% of oracle performance.

关键词： pricing data-driven algorithms robust pricing value of data distributionally robust optimization conversion rate limited information randomized algorithms

来源：评论

学校读者我要写书评

暂无评论

Automation of Photonic Networks Using Machine Learning: Case Studies and Future Works

引用

IEEE PHOTONICS TECHNOLOGY LETTERS 2021年第23期33卷 1317-1320页

作者： Rahman, Sabidur Seixas, Nilton F. S. Naznin, Mahmuda Figueiredo, Gustavo B. Sonoma State Univ Dept Comp Sci Rohnert Pk CA 94928 USA Univ Fed Bahia Dept Comp Sci BR-40170110 Salvador BA Brazil Bangladesh Univ Engn & Technol Dept Comp Sci & Engi Dhaka 1205 Bangladesh

Although a "Self-Driving" photonic network is still a long way to go, many time-consuming complex tasks and decision making in photonic networks can be automated using machine learning, and other data-driven solutions. This study explores recent contributions towards photonic network automation, such as alarm prediction, fault localization, resource auto-scaling, quality of transmission prediction, dynamic controller placement, automated service restoration, resource allocation and optimization, minimization of electricity and power supply cost, user data analysis, etc. The studies explored in this letter, provide solution approaches to these problems using wide range of data-driven methods, machine learning, AI, and deep learning based methods. This study discusses different challenges involving these interesting research areas and provides directions for future research opportunities as well.

关键词： Optical fiber networks Automation Heuristic algorithms Machine learning Resource management Predictive models Reinforcement learning Photonic network automation machine learning artificial intelligence data-driven algorithms cost savings

来源：评论

学校读者我要写书评

暂无评论

Emerging evolutionary algorithm integrated with kernel principal component analysis for modeling the performance of a water treatment plant

引用

JOURNAL OF WATER PROCESS ENGINEERING 2020年 33卷

作者： Abba, S., I Quoc Bao Pham Usman, A. G. Nguyen Thi Thuy Linh Aliyu, D. S. Quyen Nguyen Quang-Vu Bach Yusuf Maitama Sule Univ Kano Dept Phys Planning Dev Kano Nigeria Natl Cheng Kung Univ Dept Hydraul & Ocean Engn Tainan 701 Taiwan Near East Univ Fac Pharm Dept Analyt Chem Near East BlvdVia Mersin 10 TR-99138 Nicosia North Cyprus Turkey Thuyloi Univ 175 Tay Son Hanoi Vietnam Newcastle Univ Dept Civil Engn Newcastle Upon Tyne Tyne & Wear England Duy Tan Univ Inst Res & Dev Danang 550000 Vietnam

Providing a robust and reliable model is essential for hydro-environmental and public health engineering perspectives, including water treatment plants (WTPs). The current research develops an emerging evolutionary data-intelligence model: extreme learning machine (ELM) integrated with kernel principal component analysis (KPCA) to predict the performance of the Tamburawa WTP in Kano, Nigeria. A traditional feed-forward neural network (FFNN) and a classical linear autoregressive (AR) models were also employed to compare the predictive performance. For this purpose, different input data with the corresponding treated pH, turbidity, total dissolve solids, and hardness as the target variables obtained from the WTP were used. The predictive models are evaluated based on the three numerical indices, namely Nash-Sutcliffe (NC), root mean squared error (RMSE) and mean absolute percentage error (MAPE). To examine the similarities and differences between the observed and predicted values, a two-dimension graphical diagram (i.e., Taylor diagram) was also utilized. The predictive results revealed the potential of KPCA-ELM, which exhibited a high level of accuracy in comparison to the single models for all the considered variables with a slight exception in terms of pH prediction. Two different model combination were built for each single (FFNN, ELM, and AR) model and KPCA algorithms (KPCA-FFNN, KPCA-ELM, and KPCA-AR). The results also depicted that both ELM and FFNN models demonstrated prediction skill and therefore, can serve as reliable models. The outcomes may contribute to the aforementioned modeling of the treated parameters and provides a reference benchmark for wastewater management and control in the Tamburawa WTP.

关键词： Principal component analysis Water treatment plant data-driven algorithms Cross-validation Kano-Tamburawa Extreme learning marchine

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：