检索结果-内蒙古大学图书馆

Dynamical multiple polynomial-based neural networks classifier realized with the aid of dropfilter and dual statistical selection

引用

engineering Applications of Artificial Intelligence 2025年 157卷

作者： Zhen Wang Sung-Kwun Oh Zunwei Fu Seok-Beom Roh Witold Pedrycz Research Center for Big Data and Artificial Intelligence Linyi University Linyi 276005 China School of Automation and Electrical Engineering Linyi University Linyi 276005 Shandong China School of Electrical & Electronic Engineering The University of Suwon 17 Wauan-gil Bongdam-eup Hwaseong-si Gyeonggi-do 18323 South Korea Department of Electronic Engineering Seokyeong University Seoul 02713 South Korea Department of Computer Science The University of Suwon 17 Wauan-gil Bongdam-eup Hwaseong-si Gyeonggi-do 18323 South Korea Department of Electrical and Computer Engineering University of Alberta Edmonton AB T6G 2R3 Canada Systems Research Institute Polish Academy of Sciences 00-901 Warsaw Poland Department of Computer Engineering Faculty of Engineering and Natural Sciences Istinye University Sariyer Istanbul Turkiye

Polynomial neural networks (PNN) have emerged as an effective regression modeling methodology in computational intelligence, relying on its interpretable polynomial nodes to fit complex nonlinear data relationships and the adaptive nature of self-organizing networks. To break the bottleneck of PNN structure in the field of multi-classification, this study designs a dynamical multiple polynomial-based neural networks (DMPNN) classifier, focusing on developing a flexible polynomial network classification methodology that enhances predictive capabilities without sacrificing the advantages of PNN structures. Our approach effectively addresses the challenges of multi-class classification with uncertain class boundaries and reduces computational complexity, which is achieved through the synergy of several proposed techniques. Three key issues underpin the proposed DMPNN: (a) The integration of PNN regression models using the one-against-all strategy can provide effective and scalable solutions to multi-class classification problems, especially for uncertain class boundary issues. (b) The dual statistical selection (DSS) approach aims to eliminate redundant inputs during data processing, reduce the computational burden, and increase the variety of neural network nodes in the model neuron selection stage. (c) The synergy of regularization methods including the ℓ2 norm-based method (ℓ2-LSM) and the DropFilter, is exploited to mitigate potential overfitting in coefficient estimation and enhance the generalization capabilities of the proposed classifier. A series of ablation experiments and parameter analysis were conducted to demonstrate the stability and reliability of the proposed model. Then, we applied DMPNN to 17 publicly available datasets and two engineering applications: Recycling of black plastic wastes and phased resolved partial discharge. The performance results show that the DMPNN model outperforms five classical classifiers and four state-of-the-art (SOTA) class

关键词：

来源：评论

学校读者我要写书评

暂无评论

Learning active spine behaviors for dynamic and efficient locomotion

arXiv

引用

arXiv 2019年

作者： Bhattacharya, Shounak Singla, Abhik Abhimanyu Dholakiya, Dhaivat Bhatnagar, Shalabh Amrutur, Bharadwaj Ghosal, Ashitava Kolathaya, Shishir Robert Bosch Centre for Cyber-Physical Systems IISc Bangalore India Department of Mechanical Engineering BITS Pilani Goa campus India Faculty of Computer Science and Automation Faculty of Electrical and Computer Engineering Faculty of Mechanical Engineering

In this work, we provide a simulation framework to perform systematic studies on the effects of spinal joint compliance and actuation on bounding performance of a 16-DOF quadruped spined robot Stoch 2. Fast quadrupedal locomotion with active spine is an extremely hard problem, and involves a complex coordination between the various degrees of freedom. Therefore, past attempts at addressing this problem have not seen much success. Deep-Reinforcement Learning seems to be a promising approach, after its recent success in a variety of robot platforms, and the goal of this paper is to use this approach to realize the aforementioned behaviors. With this learning framework, the robot reached a bounding speed of 2:1 m/s with a maximum Froude number of 2. Simulation results also show that use of active spine, indeed, increased the stride length, improved the cost of transport, and also reduced the natural frequency to more realistic values. Copyright © 2019, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Agent DEVS simulation of the evacuation process from a commercial building during a fire 12th

Agent DEVS simulation of the evacuation process from a comme...

引用

12th International Scientific and Technical Conference computer science and Information Technologies, CSIT 2017

作者： Kulakowski, Andrzej Rogala, Bartosz Faculty of Electrical Engineering Automation Control and Computer Science Kielce University of Technology Kielce Poland

ISBN: (纸本)9783319705804

The paper describes a simulation application for the process of evacuating people from a building in the event of a fire. This simulation was developed as an agent system using the DEVS formalism. The following sections show the design and implementation of such a system using the Adevs library. A series of simulation tests were then performed for different simulation scenarios. © 2018, Springer International Publishing AG.

关键词： Office buildings

来源：评论

学校读者我要写书评

暂无评论

Software's Quality-in-use Mining for User's Comments

Software's Quality-in-use Mining for User's Comments

引用

第一届计算机科学与教育信息化国际会议

作者： Ying Jiang Tianyuan Hu Yunnan Key Lab of Computer Technology Application Kunming University of Science and Technology Faculty of Information Engineering and Automation Kunming University of Science and Technology

At present, user's comments are important information sources that help users and software developers to evaluate the software's quality-in-use. However, the accumulation of user's comment information is rapid. There are some difficulties while using user's comments to evaluate software's quality-in-use. The research and practical value of mining software's quality-in-use for user's comments are becoming more and more prominent. The systematic and comprehensive review on software's quality-in-use mining for user's comments is trying to present on the basis of related researches in recent years. There are two cores in software's quality-in-use mining for user's comments, including identification of software's quality-in-use information and analysis of software's quality-in-use information. For identification of software's quality-in-use information, it is concluded that application of feature words reflecting qualityin-use, combination with language rules of quality-in-use, extraction of evaluation aspects reflecting quality-in-use and classification based on quality-in-use characteristics. Furthermore, analysis for overall usage of software and analysis for specific usage of software are summarized. Finally, the future work of software's quality-in-use mining for user's comments is discussed.

关键词： Software's Quality-in-use Mining Technology User's Comment Information Identification Information Analysis

来源：评论

学校读者我要写书评

暂无评论

Estimating red noise in quasi-periodic signals with MCMC-based Bayesian

arXiv

引用

arXiv 2020年

作者： Liang, Bo Meng, Yao Feng, Song Yang, Yunfei Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming650500 China Key Laboratory of Solar Activity National Astronomical Observatories Chinese Academy of Sciences Beijing100012 China Yunnan Key Laboratory of Computer Technology Application Kunming650500 China

Multi-parameter Bayesian inferences based on Markov chain Monte Carlo (MCMC) samples have been widely used to estimate red noise in solar period-periodic signals. To MCMC, proper priors and sufficient iterations are prerequisites ensuring the accuracy of red noise estimation. We used MCMC-based Bayesian inferences to estimate 100 groups of red noise synthesized randomly for evaluating its accuracy. At the same time, the Brooks-Gelman algorithm was employed to precisely diagnose the convergence of the Markov chains generated by MCMC. The root-mean-square error of parameter inferences to the synthetic data is only 1.14. Furthermore, we applied the algorithm to analyze the oscillation modes in a sunspot and a flare. A 70 s period is detected in the sunspot umbra in addition to 3- and 5-minute periods, and a 40 s period is detected in the flare. The results prove that estimating red noise with MCMC-based Bayesian has more high accuracy in the case of proper priors and convergence. We also find that the number of iterations increases dramatically to achieve convergence as the number of parameters grows. Therefore, we strongly recommend that when estimating red noise with MCMC-based Bayesian, different initial values must be selected to ensure that the entire posterior distribution is covered. Copyright © 2020, The Authors. All rights reserved.

关键词： Bayesian networks

来源：评论

学校读者我要写书评

暂无评论

Anomaly Judgment for Nodes Based on Agent under Cloud Environment

Anomaly Judgment for Nodes Based on Agent under Cloud Enviro...

引用

第一届计算机科学与教育信息化国际会议

作者： Yang Lei Ying Jiang Yunnan Key Lab of Computer Technology Application Kunming University of Science and Technology Faculty of Information Engineering and Automation Kunming University of Science and Technology

Under cloud computing environment, cloud computing systems often have abnormalities due to the diversity of applications and the dynamic deployment environment. Collecting node performance data is very valuable for anomaly detection. There are many nodes under cloud computing environment. Many related studies only make abnormal judgments based on the status of single node. However, the state of multi-nodes is not judged. In this paper, a method of nodes anomaly judgment(NAJ method) based on agent under cloud environment is proposed. The collected data are analyzed by using agent technology. Firstly, data conversion technology is used for constructing a unified format for data analysis. Secondly, the detection model is trained and the degree of data anomalies is analyzed through unsupervised learning method. Finally, differences between anomalous nodes and other nodes are detected. The experimental results show that the method can judge the anomaly nodes effectively.

关键词： Cloud computing node anomaly judgment agent NAJ method

来源：评论

学校读者我要写书评

暂无评论

Analysis of Automatic Code Generation Tools based on Machine Learning

Analysis of Automatic Code Generation Tools based on Machine...

引用

第一届计算机科学与教育信息化国际会议

作者： Xiaojiang Zhang Ying Jiang Zhijun Wang Yunnan Key Lab of Computer Technology Application Kunming University of Science and Technology Faculty of Information Engineering and Automation Kunming University of Science and Technology

ISBN: (纸本)9781728123097

At present, there are many researches on automatic code generation methods based on machine learning. However, there is no analysis on automatic code generation tools based on machine learning. According to the role of existing code generation tools based on machine learning in practical de velopment, this paper proposes the content and the method of analysis of automatic code generation tools based on machine learning. Based on the presented method, the corresponding analysis plug-ins API4 ACGT are implemented. Finally, three existing automatic code generation tools based on machine learning are used to generate specific codes in two programming environments. The experimental results show that the proposed method and analysis plug-ins API4 ACGT are effective.

关键词： automatic code generation tool analysis content IDE API4ACGT

来源：评论

学校读者我要写书评

暂无评论

Comparison of the Effectiveness of Various Structures of Regulators for Controlling an Object of the Sixth Order

Comparison of the Effectiveness of Various Structures of Reg...

引用

2019 International Multi-Conference on engineering, computer and Information sciences, SIBIRCON 2019

作者： Zhmud, Vadim Dimitrov, Lubomir Nosek, Jaroslav Roth, Hubert Novosibirsk State Technical University Dept. of Automation Novosibirsk Russia Technical University of Sofia Sofia Bulgaria Technical University of Liberec Faculty of Mechatronics Computer Science and Interdisciplinary Research Liberec Czech Republic University of Siegen Department of Automatic Control Engineering Siegen Germany Germany

ISBN: (纸本)9781728144016

Object control requires the calculation of a regulator. One of the most effective methods for calculating the regulator is the method of numerical optimization of the regulator. This paper discusses the solution of the problem of controlling a linear object of the sixth order, the parameters of which are such that this high order significantly and negatively affects the possibilities of successfully managing this object. Using the traditional approach to the design problem of the controller (regulator) allows us to formally solve the problem;however, the result of this solution is not satisfactory. The use of additional derivation does not give any noticeable improvement. A solution to this problem is proposed by using pseudo-local feedback, moreover, this connection is supposedly most effective if it covers one of two identical links in the object model. This allows for a smoother attenuation of the frequency response of the object in the region where the action of these two links begins to affect. For pseudo-local communication, an additional PID controller is used, the coefficients of which are also determined by the method of numerical optimization. The optimization result allows us to conclude that the presence of an integral element in this controller is optional. The calculation result of such a composite controller meets the most stringent requirements for the quality of control, since in such a system the speed is increased at least twice, the oscillations, which were about 10% in amplitude, are completely eliminated, the robustness of the system is ensured, which is verified by changing all the controller coefficients by 5% up and down, and also by changing the time constant of the element of the pseudolocal loop by 10% up and down, changes in the transition process of the system in this case remain are acceptable. © 2019 IEEE.

关键词： Feedback

来源：评论

学校读者我要写书评

暂无评论

Optimization Algorithm for Reduction the Size of Dixon Resultant Matrix:A Case Study on Mechanical Application

引用

computers, Materials & Continua 2019年第2期58卷 567-583页

作者： Shang Zhang Seyedmehdi Karimi Shahaboddin Shamshirband Amir Mosavi College of Computer and Information Technology China Three Gorges UniversityYichangChina Department of Mathematics Jouybar branchIslamic Azad UniversityJouybarIran Department for Management of Science and Technology Development Ton Duc Thang UniversityHo Chi Minh CityVietnam Faculty of Information Technology Ton Duc Thang UniversityHo Chi Minh CityVietnam Department Institute of Structural Mechanics Bauhaus University of WeimarGermany Institute of Automation Kando Kalman Faculty of Electrical EngineeringObuda University1431 BudapestHungary

In the process of eliminating variables in a symbolic polynomial system,the extraneous factors are referred to the unwanted parameters of resulting *** paper aims at reducing the number of these factors via optimizing the size of Dixon *** optimal configuration of Dixon matrix would lead to the enhancement of the process of computing the resultant which uses for solving polynomial *** do so,an optimization algorithm along with a number of new polynomials is introduced to replace the polynomials and implement a complexity ***,the monomial multipliers are optimally positioned to multiply each of the ***,through practical implementation and considering standard and mechanical examples the efficiency of the method is evaluated.

关键词： Dixon resultant matrix symbolic polynomial system elimination theory optimization algorithm computational complexity

来源：评论

学校读者我要写书评

暂无评论

A survey on set similarity search and join

引用

International Journal of Performability engineering 2018年第2期14卷 245-258页

作者： Jia, Lianyin Zhang, Lulu Yu, Guoxian You, Jinguo Ding, Jiaman Li, Mengjuan Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming650500 China College of Computer and Information Science Southwest University Chongqing400715 China Library Yunnan Normal University Kunming650500 China

Set similarity search and join operations have a wide range of applications, including e-commerce, information retrieval, bioinformatics, and so on. Although extensive techniques have been suggested for set similarity search and join, there is literature on systematically categorizing these techniques and comprehensively comparing them. To bridge this gap, this paper provides a comprehensive survey of set similarity search and join. The survey starts from the basic definitions, framework, ordering and similarity functions of set similarity search and join. Next, it discusses the main filtering techniques used in the state-of-the-art algorithms, and analyses the pros and cons of these algorithms in detail. For set similarity join algorithms, we divide them into 2 main categories based on the key underlying techniques they use: Prefix filtering based algorithms and partition based algorithms. Prefix filtering is the most dominant technique, so algorithms based on prefix filtering and their recent variants are analyzed thoroughly. Partition is also a promising technique as it can exploit comparisons between multiple partitions. Furthermore, some efforts on MapReduce based set similarity join algorithms are also discussed briefly. In the end, the open challenges and questions are presented for future pursue. © 2018 Totem Publisher, Inc. All rights reserved.

关键词： Surveys

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：