检索结果-内蒙古大学图书馆

Cost-Sensitive Online adaptive Kernel Learning for Large-Scale Imbalanced Classification

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2023年第10期35卷 10554-10568页

作者： Chen, Yingying Hong, Zijie Yang, Xiaowei South China Univ Technol Sch Software Engn Guangzhou 510006 Peoples R China

Imbalanced classification is a challenging task in the fields of machine learning, data mining and pattern recognition. Cost-sensitive online algorithms are very important methods for large-scale imbalanced classification problems. At present, most of the cost-sensitive classification algorithms focus on the accuracy of the minority class and ignore the accuracy of the majority class. In order to better balance the accuracy between the minority class and the majority class, in this article, a misclassification cost is presented to ensure that the cost-sensitive online algorithm can better deal with the imbalanced classification problems without signifificantly reducing the accuracy of the majority class. Based on the proposed misclassification cost, a novel cost-sensitive online adaptive kernel learning algorithm is proposed to boost the adaptability of kernel function when data arrives one by one. According to the essential characteristics of the imbalanced binary classification, a cost-sensitive online adaptive kernel learning algorithm is given to handle the large-scale imbalanced multi-class classification problems. Theoretical analysis of the proposed algorithms are provided. Extensive experiments demonstrate that compared with the state-of-the-art imbalanced classification algorithms, the proposed algorithms can significantly improve the classification performances on most of the large-scale imbalanced data sets.

关键词： Classification algorithms Kernel Costs Adaptation models Machine learning algorithms Approximation algorithms Task analysis adaptive algorithms cost function classification algorithms kernel machine learning algorithms

来源：评论

学校读者我要写书评

暂无评论

A fast adaptive algorithm for training deep neural networks

引用

APPLIED INTELLIGENCE 2023年第4期53卷 4099-4108页

作者： Gui, Yangting Li, Dequan Fang, Runyue Anhui Univ Sci & Technol Sch Math & Big Data Huainan 232000 Anhui Peoples R China Anhui Univ Sci & Technol Sch Artificial Intelligence Huainan 232000 Anhui Peoples R China

Among the adaptive algorithms, Adam is the most widely used algorithm, especially for training deep neural networks. However, recent studies have shown that it has a weak generalization ability, and even cannot converge in extreme cases. AdaX (2020) is a variant of Adam, which modifies the second moment of Adam, making the algorithm enjoy good generalization ability compared to SGD. This work aims to improve the AdaX algorithm with faster convergence speed and higher training accuracy. The first moment of AdaX is essentially a classical momentum term, while the Nesterov's accelerated gradient (NAG) is theoretically and experimentally superior to this classical momentum. Therefore, we replace the classical momentum term of the first moment of AdaX with NAG, and obtain the resulting algorithm named Nesterov's accelerated AdaX (Nadax). Extensive experiments on deep learning tasks show that training models with our proposed Nadax can bring favorable benefits.

关键词： Nesterov momentum adaptive algorithms Image classification Deep nerual network

来源：评论

学校读者我要写书评

暂无评论

Performance Evaluation of Some adaptive Task Allocation algorithms for Fog Networks 5

Performance Evaluation of Some Adaptive Task Allocation Algo...

引用

5th IEEE International Conference on Fog and Edge Computing (ICFEC)

作者： Stypsanelli, Ioanna Brun, Olivier Prabhu, Balakrishna J. Univ Toulouse CNRS LAAS Toulouse France

ISBN: (纸本)9781665402910

Fog Computing brings resources closer to the end-user and improves user experience. Tasks with stringent QoS requirements can he processed locally in the Edge while the more elastic ones can be sent to the Cloud. For the benefits of this flexible architecture to he seen, task allocation algorithms should he dynamic and adapt to the load in the Fog and in the (loud. Using a discrete-event simulation approach, we evaluate the performance the four simple adaptive algorithms based on congestion estimation and compare them with the standard nearest node algorithm that uses non adaptive routing. We consider a setting in which base stations (access nodes) forward traffic to computing nodes (Fog and Cloud nodes) in a distributed way without coordination and sharing of state-information between the access and computing nodes. The algorithms are tested for their adaptability to sudden changes in the arrival rate of requests (to model peak hours) as well as robustness to the variance of the request-size distributions to understand the advantages and drawbacks of each of them. They are shown to perform well in scenarios with and without offloading.

关键词： adaptive algorithms OMNeT plus multi-armed bandits task allocation offloading

来源：评论

学校读者我要写书评

暂无评论

Recursive Constrained adaptive Filtering Algorithm Based on Arctangent Framework

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS 2023年第4期70卷 1650-1654页

作者： Jia, Wenyi Feng, Zefan Cai, Tianfu Li, Mingyu Shi, Weimin Dai, Zhijiang Chongqing Univ Sch Microelect & Commun Engn Chongqing 400044 Peoples R China

In this brief, a robust constrained filtering algorithm is proposed by introducing a novel cost function framework into the constrained adaptive algorithm. The proposed algorithm is called the recursive constrained least arctangent (RCLA) adaptive algorithm. Thanks to the robustness of arctangent function, the proposed RCLA algorithm shows superior convergence performance and better steady-state behavior against impulsive noises compared to other existing recursive methods. The mean square convergence analysis and theoretical transient mean square deviation (MSD) are derived in detail. Besides, to validate the theoretical analysis, the computer simulations are conducted to demonstrate the consistency between theoretical and simulated MSD results. Simulation results under non-Gaussian environments verify the superior behavior of the proposed RCLA algorithm compared to known algorithms.

关键词： Optimized production technology Cost function Convergence adaptive algorithms Robustness Filtering algorithms Transient analysis Recursive constrained adaptive algorithm arctangent function non-Gaussian noise convergence analysis

来源：评论

学校读者我要写书评

暂无评论

adaptive loitering anomaly detection based on motion states

引用

Multimedia Tools and Applications 2024年 1-25页

作者： Li, Hongjun Huang, Xiezhou School of Information Science and Technology Nantong University 9 Seyuan Road Jiangsu Province Nantong226019 China State Key Laboratory of Novel Software Technology Nanjing University Jiangsu Nanjing210023 China

In the field of video surveillance security in public places, loitering anomaly detection is a very critical part. Currently, the detection targets often face difficulties in loitering anomaly detection due to the complexity of the public scene, the density of pedestrians, or the varied motion states of the detection targets. This paper designs a self-adaptive loitering detection model based on motion states, and proposes a tracking method that combines local features with spatial information to enhance the correlation between adjacent frames of the same target. This model effectively reduce the impact of interleaving, occlusion, and out-of-frame situations that often occur in public places. In the detector part, a model combining frame-by-frame queue storage and self-adaptive loitering detection algorithm based on motion states is designed to solve the problem of frequent changes in motion states leading to the failure to detect loitering anomalies and to more accurately locate loitering abnormal frames. The loitering anomaly detection algorithm in this paper was tested on the IITB-Corridor dataset. Compared with the Learning regularity in skeleton trajectories (LRST) model the accuracy is improved by 4.7%, the recall is improved by 5.6%, and the precision is improved by 0.2%. The overall performance of loitering anomaly detection is improved. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： adaptive algorithms

来源：评论

学校读者我要写书评

暂无评论

Artificial Intelligence Tools for Wind Turbine Blade Monitoring

Artificial Intelligence Tools for Wind Turbine Blade Monitor...

引用

Intelligent Systems Conference

作者： Lam, Yat Ping Simani, Silvio Shibaura Inst Technol Toyosu Campus Koto City Tokyo 1358548 Japan Univ Ferrara Dept Engn Ferrara Italy

ISBN: (纸本)9783031664304;9783031664311

Electricity from wind turbines is popular and ecologically friendly. These gadgets must be reliable owing to the extensive usage of innovative materials. Researchers are creating efficient and cost-effective monitoring solutions for wind turbine blades, the most expensive part of a wind turbine. This study introduces a deep convolutional neural network-based wind turbine blade monitoring system based on medical auscultation. The system balances engineering dependability with economic efficiency. A lightweight architecture for monitoring wind turbine blades using edge computing and programmable logic controller signals is described in this study. Aerodynamic acoustic waves are collected and filtered by this technology. Our audio enhancement approaches combine self-adaptive mask targeting, multi-scale feature extraction, and deep neural networks to reduce wind turbine blade audio signal noise. Finally, we provide a unique technique to compress deep convolutional neural networks for peripheral computing devices with limited resources. Additionally, we optimise audio-generated spectrograms for wind turbine blade trouble diagnosis.

关键词： Wind turbine blade Condition monitoring Aerodynamic audio signals Must-scale feature extraction Deep neural networks adaptive algorithms

来源：评论

学校读者我要写书评

暂无评论

adaptive algorithms for fixed-cost multi-armed bandit problems with budget constraints

Adaptive algorithms for fixed-cost multi-armed bandit proble...

引用

AAMAS 2014 Workshop on adaptive and Learning Agents, ALA 2014

作者： Sen, Sandip Ridgway, Anton Department of Computer Science University of Tulsa United States

ISBN: (纸本)9781943580125

The classical multi-armed bandit problem involves pulling multiple arms with stochastic rewards with the goal of maximizing the total reward generated from those arms. A number of reinforcement learning techniques are predicated on alternate approaches to solving the exploration versus exploitation dilemma underlying this core problem. Recent work on applying this scenario to various online applications have worked on a budget-constrained version of the problem in which each arm has an associated, fixed or variable, cost and there is an assigned budget. The goal of the agent is to maximize the expected reward from pulling arms where the associated costs come from the assigned budget. We address the fixed arm pulling cost variation of the problem with several adaptive arm-selection strategies that progressively eliminate arms that are found to be less rewarding. We argue for the use of such conservative arm-elimination schemes over previously developed aggressive elimination schemes that select the best arm after a predetermined exploration phase. We also demonstrate the advantage of "forgiving" approaches that can revisit previously eliminated arms and show that those variations improve all algorithms studied for this problem.

关键词： adaptive algorithms

来源：评论

学校读者我要写书评

暂无评论

Communication Cost of Joins over Federated Data 27

Communication Cost of Joins over Federated Data

引用

27th International Conference on Database Theory (ICDT)

作者： Cucumides, Tamara Reutter, Juan Univ Antwerp Antwerp Belgium PUC Chile Santiago Chile IMFD Chile Santiago Chile

ISBN: (纸本)9783959773126

We study the problem of querying different data sources, which we assume out of our control and that are made available by standard web communication protocols. In this scenario, the time spent communicating data often dominates the time spent processing local queries in each server. Thus, our focus is on algorithms that minimize the communication between the query processing server and the federated servers containing data. However, any federated query can always be answered with linear communication, simply by requesting all the data to the federated sources. Further, one can show that certain queries do require this amount of communication. But sending all the data is definitely not a relevant algorithm from a practical point of view. This worst-case analysis is, therefore, not useful for our needs. There is a growing body of work in terms of designing strategies that minimize communication in query federation, but these strategies are commonly based in heuristics, and we currently miss a formal analysis providing guidelines for the design of such strategies. We focus on the communication complexity of federated joins when the problem is parameterized by a measure commonly referred to as the certificate of the instance: a framework that has been used before in the context of set intersection and local query processing. We show how to process any conjunctive query in time given by the certificate of instances. Our algorithm is an adaptation of Minesweeper, one of the algorithms devised for local query processing, into our federating setting. When certificates are of the size of the instance, this amount to sending the entire database, but our strategy provides drastic reductions in the communication needed for queries and instances with small certificates. We also show matching communication lower bounds for cases where the certificate is smaller than the size of active domain of the instances.

关键词： databases database queries query federation communication complexity adaptive algorithms

来源：评论

学校读者我要写书评

暂无评论

Smart Strategies for Project Scheduling: An adaptive BR-DEH Approach 2nd

Smart Strategies for Project Scheduling: An Adaptive BR-DEH ...

引用

2nd Decision Science Alliance International Summer Conference-DSA ISC

作者： Martin, Xabier A. Panadero, Javier Juan, Angel A. Univ Politecn Valencia Res Ctr Prod Management & Engn Plaza Ferrandiz Salvador Alcoy 03801 Spain Univ Autonoma Barcelona Dept Comp Architecture & Operating Syst Carrer Sitges Bellaterra 08193 Spain

ISBN: (纸本)9783031782404;9783031782411

In recent years, management of multiple complex projects has become increasingly common among large-scale companies, posing significant challenges for company managers. These projects are often identified by numerous tasks that have precedence relations and share a set of limited resources. The main goal of project scheduling is to minimize the total time required to complete these projects. However, this requires careful consideration of resource allocation, careful scheduling choices, and exact sequencing to maximize efficiency while managing the complex interactions between task dependencies. A novel discrete-event heuristic is presented to solve this project scheduling problem, which is later extended into a probabilistic algorithm using biased-randomization techniques. In addition, an adaptive mechanism was introduced to tune the parameters of the algorithm for optimal scheduling in different problem instances. Computational results demonstrate the effectiveness of our approach, finding high-quality solutions for difficult problem sets in short computational times. This intuitive and efficient approach can empower company managers to efficiently improve operational efficiency, reduce project costs, and ultimately increase their business's success.

关键词： Project Scheduling Discrete-Event Heuristic adaptive algorithms

来源：评论

学校读者我要写书评

暂无评论

A Review of Recent Advances in Discretization Methods, a Posteriori Error Analysis, and adaptive algorithms for Numerical Modeling in Geosciences

引用

OIL & GAS SCIENCE AND TECHNOLOGY-REVUE D IFP ENERGIES NOUVELLES 2014年第4期69卷 701-729页

作者： Di Pietro, Daniele A. Vohralik, Martin Univ Montpellier 2 I3M F-34057 Montpellier 5 France INRIA Paris Rocquencourt F-78153 Le Chesnay France

Two research subjects in geosciences which lately underwent significant progress are treated in this review. In the first part, we focus on one key ingredient for the numerical approximation of the Darcy flow problem, namely the discretization of diffusion terms on general polygonal/polyhedral meshes. We present different schemes and discuss in detail their fundamental numerical properties such as stability, consistency, and robustness. The second part of the paper is devoted to error control and adaptivity for model problems in geosciences. We present the available a posteriori estimates guaranteeing the maximal overall error and show how the different error components can be identified. These estimates are used to formulate adaptive stopping criteria for linear and nonlinear solvers, time step choice adjustment, and adaptive mesh refinement. Numerical experiments illustrate such entirely adaptive algorithms.

关键词： Discretization numerical approximation Earth sciences adaptive algorithms

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：