检索结果-内蒙古大学图书馆

A Hub-Based self-organizing algorithm for Feedforward Small-World Neural Network

IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE 2025年第1期9卷 160-175页

作者： Li, Wenjing Chen, Can Qiao, Junfei Beijing Univ Sci & Technol Sch Informat Engn Beijing 100124 Peoples R China Beijing Artificial Intelligence Inst Beijing 100124 Peoples R China Beijing Key Lab Computat Intelligence & Intelligen Beijing 100124 Peoples R China Beijing Lab Intelligent Environm Protect Beijing 100124 Peoples R China

By integrating the small-world (SW) property into the design of feedforward neural networks, the network performance would be improved by well-documented evidence. To achieve the structural self-adaptation of the feedforward small-world neural networks (FSWNNs), a self-organizing FSWNN, namely SOFSWNN, is proposed based on a hub-based self-organizing algorithm in this paper. Firstly, an FSWNN is constructed according to Watts-Strogatz's rule. Derived from the graph theory, the hub centrality is calculated for each hidden neuron and then used as a measurement for its importance. The self-organizing algorithm is designed by splitting important neurons and merging unimportant neurons with their correlated neurons, and the convergence of this algorithm can be guaranteed theoretically. Extensive experiments are conducted to validate the effectiveness and superiority of SOFSWNN for both classification and regression problems. SOFSWNN achieves an improved generalization performance by SW property and the self-organizing structure. Besides, the hub-based self-organizing algorithm would determine a compact and stable network structure adaptively even from different initial structure.

关键词： Feedforward neural networks small-world property hub centrality self-organizing algorithm self-organizing algorithm small-world property hub centrality self-organizing algorithm

来源：评论

学校读者我要写书评

暂无评论

self-organizing algorithm of robust PCA based on single-layer NN

Self-organizing algorithm of robust PCA based on single-laye...

引用

4th International Conference on Document Analysis and Recognition (ICDAR 97)

作者： Wang, S Xia, SW Department of Automation Tsinghua University Beijing China

ISBN: (纸本)0818678984;0818678992

One way to improve the robustness of principal component analysis (PCA) is studied in order to increase the accuracy of feature-extraction in bank-check recognition. The typical two kinds of aspects to analyze the robustness of PCA algorithm are proposed and compared: one is based on the independent among the acquired principal components and the other is based on reducing the effects of the outliers in the training sample set. A new self-organizing algorithm of robust PCA is presented based on the structure of single-layer neural network (NN) with the modification of the cost function which stands for the reconstruction error of the input signal. The new nonlinear robust PCA algorithm can recognize the outliers in the training sample set automatically and exterminate their effects to the accuracy and convergence of the PCA algorithm through proper processing to the recognized outliers.

关键词： neural nets feature extraction document image processing cheque processing bank data processing convergence of numerical methods self-adjusting systems algorithm theory image recognition self-organizing algorithm robust principal component analysis feature extraction bank check recognition outliers training sample set single-layer neural network cost function input signal reconstruction error accuracy convergence

来源：评论

学校读者我要写书评

暂无评论

A multistage self-organizing algorithm combined transiently chaotic neural network for cellular channel assignment

引用

IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 2002年第6期51卷 1386-1396页

作者： He, ZY Zhang, YF Wei, CJ Wang, J Southeast Univ Dept Radio Engn Nanjing 210096 Peoples R China Chinese Univ Hong Kong Dept Automat & Comp Aided Engn Shatin Hong Kong Peoples R China

In this paper, a new multistage self-organizing channel assignment algorithm with a transiently chaotic neural network (MSSO-TCNN) is proposed as an optimization algorithm. The algorithm is used for assigning channels in cellular mobile networks to cells in the frequency domain. The MSSO-TCNN consists of a progressively initial channel assignment stage and the TCNN assignment stage. According to the difficulty measure of each cell, the first stage is executed to assign channels cell by cell inspired by the mechanism of bristle. If the optimum assignment solution is not obtained in the first stage, the TCNN stage is then applied to continue the channel assignment until the optimum assignment is made or a maximum number of iterations is reached. A salient feature of the TCNN model is that chaotic neurodynamics are temporarily generated for searching and self-organizing in order to escape local minima. Therefore, the neural network gradually approaches, through transient chaos, a dynamical structure similar to conventional models such as the Hopfield neural network and converges to a stable equilibrium point. A variety of testing problems are used to compare the performance of the MSSO-TCNN against existing heuristic approaches. Simulation results show that the MSSO-TCNN improves performance substantially through solving well-known benchmark problems within comparable numbers of iterations to most existing algorithms.

关键词： cellular systems channel assignment chaos neural networks self-organizing algorithm

来源：评论

学校读者我要写书评

暂无评论

Fault-Tolerant Control for Robotic Systems Using a Wavelet Type-2 Fuzzy Brain Emotional Learning Controller and a TOPSIS-Based self-organizing algorithm

引用

INTERNATIONAL JOURNAL OF FUZZY SYSTEMS 2023年第5期25卷 1727-1741页

作者： Pham, Duc-Hung Huynh, Tuan-Tu Lin, Chih-Min Yuan Ze Univ Dept Elect Engn Taoyuan 320 Taiwan Hung Yen Univ Technol & Educ Fac Elect & Elect Engn Hung Yen 160000 Vietnam Lac Hong Univ Fac Mechatron & Elect Bien Hoa 810000 Vietnam

To improve control performance in robotic systems with faults and unknown uncertainties, a fault-tolerant control (FTC) system based on a wavelet type-2 fuzzy brain emotional learning controller (WT2FBC) is developed in this research. Type-2 fuzzy logic systems can handle uncertainties better than type-1 fuzzy logic systems, so they can solve some uncertain complex problems more accurate. Given its fast learning and problem-solving capabilities, the proposed WT2FBC is an important fuzzy neural network. The next step is to investigate how a WT2FBC-based FTC system can help solve the puzzles that plague the robotic systems. To achieve FTC, a control system with a computed torque controller and the WT2FBC estimator is provided. If the robotic system ever exhibits faults, it can consult an online fault estimator built on WT2FBC, and then compensates the faults. For more efficient, the total number of fuzzy rules of WT2FBC can be dynamically adjusted using a self-organizing algorithm based on the technique for order of preference by similarity to ideal solution. Simulations of a biped robot and a two-link manipulator robot have demonstrated the effectiveness of the proposed control algorithm.

关键词： Wavelet type-2 fuzzy system Brain emotional learning controller self-organizing algorithm Fault-tolerant control Biped robot Two-link manipulator robot

来源：评论

学校读者我要写书评

暂无评论

An online self-organizing algorithm for feedforward neural network

引用

NEURAL COMPUTING & APPLICATIONS 2020年第23期32卷 17505-17518页

作者： Qiao, Jun-fei Guo, Xin Li, Wen-jing Beijing Univ Technol Fac Informat Technol Beijing 100124 Peoples R China Beijing Key Lab Computat Intelligence & Intellige Beijing 100124 Peoples R China

Feedforward neural network (FNN) is the most popular network model, and the appropriate structure and learning algorithms are the key of its performance. This paper proposes an online self-organizing algorithm for feedforward neural network (OSNN) with a single hidden layer. The proposed OSNN optimizes the structure of FNN for time-varying system including structure design and parameter learning. In structure design, this paper measures the contribution ratios of hidden nodes by local sensitivity analysis based on differentiation method. OSNN merges hidden nodes with the others that have the highest correlation when their contribution ratios are almost zero and adds new hidden nodes by error reparation. For parameter learning, an improved online gradient method (OGM), called online gradient method with fixed memory (FMOGM), is proposed to improve the convergence speed and accuracy of OGM. In addition, this paper calculates the contribution ratios and the network error and estimates the local minima by using the fixed-sized training set of FMOGM instead of one sample at the current time, which can obtain more effective local information and a compact network structure. Finally, the proposed OSNN is verified using a number of benchmark problems and a practical problem for biochemical oxygen demand prediction in wastewater treatment. The experimental results show that OSNN has better convergence speed and accuracy than other algorithms.

关键词： Feedforward neural network self-organizing algorithm Online gradient method Local sensitivity analysis

来源：评论

学校读者我要写书评

暂无评论

Optimum Design of Function-Link Type-2 Fuzzy Asymmetric CMAC Based on self-organizing algorithm and Modified Jaya algorithm

引用

IEEE ACCESS 2020年 8卷 202365-202378页

作者： Le, Tien-Loc Quynh, Nguyen Vu Long, Ngo Kim Hong, Sung Kyung Sejong Univ Fac Mech & Aerosp Seoul 05006 South Korea Lac Hong Univ Fac Mechatron & Elect Bien Hoa 810000 Vietnam

In this work, a novel self-organizing interval type-2 fuzzy cerebellar model articulation controller design for chaotic systems synchronization is presented. A function-link network and an asymmetric-membership function are employed to improve the learning ability and flexibility of the proposed network. The gradient-descent method is used to derive the adaptive laws for online updating the network parameters. Suitable learning rates for adaptive laws can be achieved by employing the modified Jaya algorithm. Furthermore, the proposed self-organizing algorithm is used to generate new fuzzy rules or delete less-active fuzzy rules. The control system's exponential stability is analyzed by applying the Lyapunov theory. Finally, numerical simulations on the synchronization of four-dimensional chaotic systems (considering external disturbances and system uncertainties) are conducted to validate the efficiency and performance of the proposed method.

关键词： self-organizing algorithm type-2 fuzzy CMAC Jaya algorithm function-link network asymmetric membership function

来源：评论

学校读者我要写书评

暂无评论

Use of Sensor Networks with self-organizing algorithm to Increase the Agricultural Productivity 5

Use of Sensor Networks with Self-Organizing Algorithm to Inc...

引用

5th IEEE International Conference on Computing, Power, and Communication Technologies, IC2PCT 2024

作者： Lahoti, Saurabh Chitkara University Institute of Engineering and Technology Chitkara University Centre for Interdisciplinary Research in Business and Technology Punjab India

ISBN: (纸本)9798350383522

Traditional farming methods are becoming incapable of keeping up with global population growth. As a result, innovative farming ideas are desperately necessary to meet the food needs of a growing population. Intelligent farming systems based on sensor networks with self-organizing algorithms have gained popularity in recent years as a means of increasing agricultural production. In this study, we will use sensor networks with self-organizing algorithms to boost agricultural productivity. Finally, the paper describes the future applications of appropriate technology in agricultural production. © 2024 IEEE.

关键词： Agricultural Productivity self-organizing algorithm Sensor Networks

来源：评论

学校读者我要写书评

暂无评论

A self-organizing clustering algorithm for functional data

引用

COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION 2020年第5期49卷 1237-1263页

作者： Chen, Jen-Hao Chang, Yen-Chang Hung, Wen-Liang Natl Tsing Hua Univ Inst Computat & Modeling Sci Hsinchu Taiwan Natl Tsing Hua Univ Ctr Gen Educ Hsinchu Taiwan Natl Tsing Hua Univ Ctr Teacher Educ Hsinchu Taiwan Natl Tsing Hua Univ Dept Comp Sci Hsinchu Taiwan

Based on a learning schema, we proposed a self-organizing clustering algorithm for functional data. The proposed algorithm can be used to cluster functional data without any prior knowledge of present clusters in the data set. The resulting clusters represent homogeneity in the input data. The theoretical analysis of the convergence of the proposed algorithm is also given. Comparisons with three well-known clustering algorithms, fuzzy k-means, k-means and Funclust, show that the proposed algorithm outperforms these competitors and provides better corrected classification rate and value of variance ratio criterion.

关键词： Clustering k-means Functional data self-organizing algorithm

来源：评论

学校读者我要写书评

暂无评论

Feasibility Analysis and self-organizing algorithm for RAN Cooperative Caching

Feasibility Analysis and Self-organizing Algorithm for RAN C...

引用

IEEE Wireless Communications and Networking Conference

作者： Zejue Wang Hongjia Li Chang Yang State Key Laboratory of Information Security Institute of Information Engineering Chinese Academy of Sciences

ISBN: (纸本)9781467398152

The radio access network (RAN) cooperative caching, which explores the scale effect through cooperative content sharing and caching among multiple RAN caches, is considered as one effective way to fully benefit from the RAN cache. In this paper, we study the feasibility and self-organizing algorithm for RAN cooperative caching. Specifically, we first analyze the real-world dataset of daily content requests from 10 LTE Base Stations (BSs), and find that pursuing high hit rate does not guarantee the reduction of backhaul traffic. Besides, it is shown that content requests from different BSs feature strongly temporal and spatial correlations. To the best of our knowledge, this finding proves the feasibility of RAN cooperative caching for the first time. Then, based on our findings, we propose the self-organizing algorithm for RAN caches to individually decide on how to update their cached content objects, utilizing the defined utility with consideration of the link states among RAN caches, and the size and the request number of the missed and cached content objects. Finally, the performance of our proposed algorithm is validated based on real-world dataset. The results show that the proposed algorithm achieves significant improvement in reducing backhaul traffic. For instance, when cache capacity of each BS is 65% of the traffic generated by non-repeated content objects which are requested in one day over its coverage, the proposed algorithm with 8 defined cooperative caches (CoCas) reaches an average reduction of more than 60% of the backhaul traffic generated in one day over the BS's coverage.

关键词： Telecommunications RAN gene Cooperative caching self-organizing algorithm scale effect Cache memory Traffic hit rate return stroke

来源：评论

学校读者我要写书评

暂无评论

A Crop Canopy Localization Method Based on Ultrasonic Ranging and Iterative self-organizing Data Analysis Technique algorithm

引用

SENSORS 2020年第3期20卷 818-818页

作者： Li, Fang Bai, Xiaohu Li, Yongkui Shenyang Agr Univ Coll Engn Shenyang 110866 Peoples R China Liaoning Shihua Univ Sch Min Engn Fushun 113001 Peoples R China

To protect crops from diseases and increase yields, chemical agents are applied by boom sprayers. To achieve the optimal effect, the boom and the crop canopy should be kept at an appropriate distance. So, it is crucial to be able to distinguish the crop canopy from other plant leaves. Based on ultrasonic ranging, this paper adopts the fuzzy iterative self-organizing data analysis technique algorithm to identify the canopy location. According to the structural characteristics of the crop canopy, based on fuzzy clustering, the algorithm can dynamically adjust the number and center of clusters so as to get the optimal results. Therefore, the distances from the sensor to the canopy or the ground can be accurately acquired, and the influence of lower leaves on the measurement results can be alleviated. Potted corn plants from the 3-leaf stage to the 6-leaf stage were tested on an experiment bench. The results showed that the calculated distances from the sensor to the canopy using this method had good correlation with the manually measured distances. The maximum error of calculated values appeared at the 3-leaf stage. With the growth of plants, the error of calculated values decreased. The increased sensor moving speeds led to increased error due to the reduced data points. From the 3-leaf stage to the 5-leaf stage, the distances from the sensor to the ground can also be obtained at the same time. The method proposed in this paper provides a practical resolution to localize the canopy for adjusting the height of sprayer boom.

关键词： canopy localization fuzzy clustering self-organizing algorithm ultrasonic sensor sprayer boom height control

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：