Pub Date : 2021-12-17DOI: 10.1109/PIC53636.2021.9686999
Thanasis Kotsiopoulos, T. Vafeiadis, Aristeidis Apostolidis, Alexandros Nizamis, Nikolaos Alexopoulos, D. Ioannidis, D. Tzovaras, P. Sarigiannidis
In this work a comparative study among the known fault detection techniques Local Outlier Factor and Isolation Forest as well as a proposed methodology called Standardised Mahalanobis Distance is presented. The study is focusing on the challenging problem of fault detection on bearings and rotating machines using vibration sensors’ data. During the first phase of the experiments, all models are applied and evaluated using cross-validation on a dataset created in lab by obtaining vibration signals of a rotating machine. In the second phase, the outlier detection techniques including the proposed one, are applied and evaluated on a popular, public dataset. In both phases, various parameters’ combinations are tested in order to find the most efficient set for each technique. As can been derived by the evaluation results, the Standardised Mahalanobis Distance methodology outperforms Local Outlier Factor and Isolation Forest on fault detection on voltage drop down of rotating machines in the case the voltage value of the abnormal condition is not close to the nominal. In addition, the evaluation results from the public dataset indicate that Standardised Mahalanobis Distance is able to identify outliers before an outer race fault on a bearing occurs, in a more efficient and solid way than Local Outlier Factor and Isolation Forest models. The proposed approach is applied also on a real world scenario in the premises of major lift manufacturer, using custom vibration sensors and it is currently under further evaluation.
{"title":"Fault Detection on Bearings and Rotating Machines based on Vibration Sensors Data","authors":"Thanasis Kotsiopoulos, T. Vafeiadis, Aristeidis Apostolidis, Alexandros Nizamis, Nikolaos Alexopoulos, D. Ioannidis, D. Tzovaras, P. Sarigiannidis","doi":"10.1109/PIC53636.2021.9686999","DOIUrl":"https://doi.org/10.1109/PIC53636.2021.9686999","url":null,"abstract":"In this work a comparative study among the known fault detection techniques Local Outlier Factor and Isolation Forest as well as a proposed methodology called Standardised Mahalanobis Distance is presented. The study is focusing on the challenging problem of fault detection on bearings and rotating machines using vibration sensors’ data. During the first phase of the experiments, all models are applied and evaluated using cross-validation on a dataset created in lab by obtaining vibration signals of a rotating machine. In the second phase, the outlier detection techniques including the proposed one, are applied and evaluated on a popular, public dataset. In both phases, various parameters’ combinations are tested in order to find the most efficient set for each technique. As can been derived by the evaluation results, the Standardised Mahalanobis Distance methodology outperforms Local Outlier Factor and Isolation Forest on fault detection on voltage drop down of rotating machines in the case the voltage value of the abnormal condition is not close to the nominal. In addition, the evaluation results from the public dataset indicate that Standardised Mahalanobis Distance is able to identify outliers before an outer race fault on a bearing occurs, in a more efficient and solid way than Local Outlier Factor and Isolation Forest models. The proposed approach is applied also on a real world scenario in the premises of major lift manufacturer, using custom vibration sensors and it is currently under further evaluation.","PeriodicalId":297239,"journal":{"name":"2021 IEEE International Conference on Progress in Informatics and Computing (PIC)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132308771","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-17DOI: 10.1109/PIC53636.2021.9687073
Chang Lu, Dayu Feng, Jie Zhu, Haiping Huang
As a NP-hard problem, it is always baffling to figure out a scheduling strategy to arrange the interconnected tasks of a workflow on the infinite number of resources in the cloud environment so that the workflow can be addressed efficiently and robustly. This paper focuses on scheduling the workflow’s tasks on the cloud resources with less rental cost of resources while the whole schedule length (makespan) will not exceed the given deadline. As one of the most popular evolutionary algorithms, particle swarm optimization (PSO) has been successfully applied for the workflow scheduling problem. Inspired by the idea of multiple groups and the distributed parallel computing, we develop an enhanced PSO algorithm for the workflow scheduling problem in clouds. Besides, a pretreatment strategy is adopted to simplify the workflow’s structure. The experimental results demonstrate that our proposal has good performance on improving the algorithm’s searching ability and finding better solutions.
{"title":"Enhanced Particle Swarm Optimization for Workflow Scheduling in Clouds","authors":"Chang Lu, Dayu Feng, Jie Zhu, Haiping Huang","doi":"10.1109/PIC53636.2021.9687073","DOIUrl":"https://doi.org/10.1109/PIC53636.2021.9687073","url":null,"abstract":"As a NP-hard problem, it is always baffling to figure out a scheduling strategy to arrange the interconnected tasks of a workflow on the infinite number of resources in the cloud environment so that the workflow can be addressed efficiently and robustly. This paper focuses on scheduling the workflow’s tasks on the cloud resources with less rental cost of resources while the whole schedule length (makespan) will not exceed the given deadline. As one of the most popular evolutionary algorithms, particle swarm optimization (PSO) has been successfully applied for the workflow scheduling problem. Inspired by the idea of multiple groups and the distributed parallel computing, we develop an enhanced PSO algorithm for the workflow scheduling problem in clouds. Besides, a pretreatment strategy is adopted to simplify the workflow’s structure. The experimental results demonstrate that our proposal has good performance on improving the algorithm’s searching ability and finding better solutions.","PeriodicalId":297239,"journal":{"name":"2021 IEEE International Conference on Progress in Informatics and Computing (PIC)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124274636","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-17DOI: 10.1109/PIC53636.2021.9687088
Qing Li, Jun Zhang
Photon-limited Poisson image denoising is a pressing problem and faces great challenges in some fields such as emission tomography, low-exposure x-ray imaging, fluorescence microscopy, and infrared astronomy. Currently, the post- processing best linear prediction method (BLP) based on co- variance estimation of non-local similar image patches has been proposed and achieved good results in Poisson image denoising. However, the calculation of similarity is inaccurate in the photon limited case, which leads to the inaccuracy of similarity patches- based covariance estimation as well. To remedy this, we propose a new BLP method based on weighted covariance estimation (WBLP). This method searches for similar patches in a large window for each reference patch, which brings a large amount of computation. To solve this problem, we introduce a randomized acceleration technique to speed up our method.
{"title":"Weighted Best Linear Prediction and Its Randomized Acceleration for Poisson Image Denoising","authors":"Qing Li, Jun Zhang","doi":"10.1109/PIC53636.2021.9687088","DOIUrl":"https://doi.org/10.1109/PIC53636.2021.9687088","url":null,"abstract":"Photon-limited Poisson image denoising is a pressing problem and faces great challenges in some fields such as emission tomography, low-exposure x-ray imaging, fluorescence microscopy, and infrared astronomy. Currently, the post- processing best linear prediction method (BLP) based on co- variance estimation of non-local similar image patches has been proposed and achieved good results in Poisson image denoising. However, the calculation of similarity is inaccurate in the photon limited case, which leads to the inaccuracy of similarity patches- based covariance estimation as well. To remedy this, we propose a new BLP method based on weighted covariance estimation (WBLP). This method searches for similar patches in a large window for each reference patch, which brings a large amount of computation. To solve this problem, we introduce a randomized acceleration technique to speed up our method.","PeriodicalId":297239,"journal":{"name":"2021 IEEE International Conference on Progress in Informatics and Computing (PIC)","volume":"228 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121465310","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-17DOI: 10.1109/PIC53636.2021.9687067
Cong Liu, Xiaobin Xu, Xinhong Li, Zhijie Pan, Kai Hu, You Shu
A path planning framework for an omnidirectional mobile robot based on modified A * algorithm with energy model is proposed. The kinematic model of the omnidirectional mobile robot is established, and the energy consumption model of the omnidirectional mobile robot during motion is established based on the detailed derivation of formulas and used to improve the A* algorithm. The derived results show that the extra power consumed in steering movement is only related to the steering angle.
{"title":"Path Planning for an Omnidirectional Mobile Robot Based on Modified A * Algorithm with Energy Model","authors":"Cong Liu, Xiaobin Xu, Xinhong Li, Zhijie Pan, Kai Hu, You Shu","doi":"10.1109/PIC53636.2021.9687067","DOIUrl":"https://doi.org/10.1109/PIC53636.2021.9687067","url":null,"abstract":"A path planning framework for an omnidirectional mobile robot based on modified A * algorithm with energy model is proposed. The kinematic model of the omnidirectional mobile robot is established, and the energy consumption model of the omnidirectional mobile robot during motion is established based on the detailed derivation of formulas and used to improve the A* algorithm. The derived results show that the extra power consumed in steering movement is only related to the steering angle.","PeriodicalId":297239,"journal":{"name":"2021 IEEE International Conference on Progress in Informatics and Computing (PIC)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124852135","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The problem of job shop scheduling is a hot research topic nowadays. How to improve the production efficiency of the equipment and shorten the processing time of the workpieces has become an important research work. The parallelism and mechanism of distributed computing of Ant colony optimization (ACO) provide a good solution in solving job shop scheduling problems. In this paper, the ACO is applied to the job shop scheduling of industrial production. And the ACO is used to solve the scheduling problem, the pheromone update strategy in the ant colony algorithm has been modified, and roulettes wheel was introduced. On the basis of above modifications, a job shop scheduling method based on ant colony algorithm has been used in this paper. In addition, the disjunction graph model of the job shop problem has been also established in this paper, which turned the job shop scheduling problem into a solution to the traveling salesman problem and then redefined as a natural expression model suitable for ant colony algorithm. When solving the traveling salesman problem, virtual nodes were added as the super source and destination in the search process, the distance between cities and the shortest path in the traveling salesman were corresponded with the processing time and the shortest processing time in the job shop scheduling problem one by one. In this paper, C++ has been used for programming, and the FT06 data example was used as a test example. In the experiment, the scheme of job scheduling with minimum total completion time was obtained successfully, which verified the feasibility and effectiveness of this method in the shop scheduling problem.
{"title":"A Job Shop Scheduling Method Based on Ant Colony Algorithm","authors":"Junqing Li, Huawei Deng, Dawei Liu, Changqing Song, Ruiyi Han, Taiyuan Hu","doi":"10.1109/PIC53636.2021.9687078","DOIUrl":"https://doi.org/10.1109/PIC53636.2021.9687078","url":null,"abstract":"The problem of job shop scheduling is a hot research topic nowadays. How to improve the production efficiency of the equipment and shorten the processing time of the workpieces has become an important research work. The parallelism and mechanism of distributed computing of Ant colony optimization (ACO) provide a good solution in solving job shop scheduling problems. In this paper, the ACO is applied to the job shop scheduling of industrial production. And the ACO is used to solve the scheduling problem, the pheromone update strategy in the ant colony algorithm has been modified, and roulettes wheel was introduced. On the basis of above modifications, a job shop scheduling method based on ant colony algorithm has been used in this paper. In addition, the disjunction graph model of the job shop problem has been also established in this paper, which turned the job shop scheduling problem into a solution to the traveling salesman problem and then redefined as a natural expression model suitable for ant colony algorithm. When solving the traveling salesman problem, virtual nodes were added as the super source and destination in the search process, the distance between cities and the shortest path in the traveling salesman were corresponded with the processing time and the shortest processing time in the job shop scheduling problem one by one. In this paper, C++ has been used for programming, and the FT06 data example was used as a test example. In the experiment, the scheme of job scheduling with minimum total completion time was obtained successfully, which verified the feasibility and effectiveness of this method in the shop scheduling problem.","PeriodicalId":297239,"journal":{"name":"2021 IEEE International Conference on Progress in Informatics and Computing (PIC)","volume":"124 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127260600","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-17DOI: 10.1109/PIC53636.2021.9687039
Zicheng Zhang, Quan Liang, Zhihui Feng, W. Ji, Hansong Wang, Jinjing Hu
With the development of unmanned technology, the technical innovation of invehicle vision detection system is also getting faster and faster, while the improvement of algorithm accuracy often brings an increase in the number of parameters and poor real-time performance. In his paper, the optimization of the algorithm structure of YoLoV4 target detection is achieved by using MobileNet-v3 instead of the CspDarkNet53 master Network, which has the inverse residual structure of linear bottleneck, while the lightweight attention mechanism is added to the feature extraction process, and the learning degree of feature channels is enhanced; due to the long computation time of sigmoid, it also uses ReLU6(x+3)/6 is used to approximate the original activation function due to the long computation time of sigmoid; the system parameters are reduced by constructing a depth-separable convolution instead of the normal convolution in PaNet. Meanwhile, this paper improves the original upsampling method by using dual cubic interpolation, which makes the image more smooth, less image loss and more accurate feature extraction during he upsampling method. The map% is improved from 79.1% to 81.2% on the voc dataset, reaching 58.14 FPS.
{"title":"Application of Improved YOLOV4 in Intelligent Driving Scenarios","authors":"Zicheng Zhang, Quan Liang, Zhihui Feng, W. Ji, Hansong Wang, Jinjing Hu","doi":"10.1109/PIC53636.2021.9687039","DOIUrl":"https://doi.org/10.1109/PIC53636.2021.9687039","url":null,"abstract":"With the development of unmanned technology, the technical innovation of invehicle vision detection system is also getting faster and faster, while the improvement of algorithm accuracy often brings an increase in the number of parameters and poor real-time performance. In his paper, the optimization of the algorithm structure of YoLoV4 target detection is achieved by using MobileNet-v3 instead of the CspDarkNet53 master Network, which has the inverse residual structure of linear bottleneck, while the lightweight attention mechanism is added to the feature extraction process, and the learning degree of feature channels is enhanced; due to the long computation time of sigmoid, it also uses ReLU6(x+3)/6 is used to approximate the original activation function due to the long computation time of sigmoid; the system parameters are reduced by constructing a depth-separable convolution instead of the normal convolution in PaNet. Meanwhile, this paper improves the original upsampling method by using dual cubic interpolation, which makes the image more smooth, less image loss and more accurate feature extraction during he upsampling method. The map% is improved from 79.1% to 81.2% on the voc dataset, reaching 58.14 FPS.","PeriodicalId":297239,"journal":{"name":"2021 IEEE International Conference on Progress in Informatics and Computing (PIC)","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114562658","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-17DOI: 10.1109/PIC53636.2021.9687057
Shitong Shen, Jian Cao, Yinyue Yang, Yameng Guo
Compared with the air transportation and land transportation, water transportation has many advantages such as larger loading capacity, lower unit transportation cost, lower construction investment and so on. What’s more, water transportation has played an important role in the economical development of China, especially in the aspect of international trade. Therefore, the improvement in the efficiency of water transportation will be of great significance. In this paper, we designed a system to predict the containers’ entering time distribution of a given voyage at a specific port by using machine learning algorithms and statistical methods. Using Shanghai Yangshan Port phase IV automated terminal’s data, we perform some experiments, and the result shows that our system can provide valid predictions.
{"title":"A kNN Based Voyage’s Containers’ Entering Time Distribution Prediction System","authors":"Shitong Shen, Jian Cao, Yinyue Yang, Yameng Guo","doi":"10.1109/PIC53636.2021.9687057","DOIUrl":"https://doi.org/10.1109/PIC53636.2021.9687057","url":null,"abstract":"Compared with the air transportation and land transportation, water transportation has many advantages such as larger loading capacity, lower unit transportation cost, lower construction investment and so on. What’s more, water transportation has played an important role in the economical development of China, especially in the aspect of international trade. Therefore, the improvement in the efficiency of water transportation will be of great significance. In this paper, we designed a system to predict the containers’ entering time distribution of a given voyage at a specific port by using machine learning algorithms and statistical methods. Using Shanghai Yangshan Port phase IV automated terminal’s data, we perform some experiments, and the result shows that our system can provide valid predictions.","PeriodicalId":297239,"journal":{"name":"2021 IEEE International Conference on Progress in Informatics and Computing (PIC)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124046224","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-17DOI: 10.1109/PIC53636.2021.9687007
X. Nan, R. Chen, Hongtao Tian, Yupeng Liu
For the question that situation assessment methods for the analysis of existing vulnerabilities are associated with the lack of analysis of vulnerability assessments, which leads to the poor accuracy assessment, the paper presents a method for network vulnerabilities associated with risk assessment situation analysis. The method improves the existing hierarchical network situation assessment, with the system being divided into three levels, which are loopholes at the bottom, host in the middle, and network system at the top. Based on the security risk indices, we calculate the vulnerability, the host, the entire network system risk index, and evaluate and analyze the security posture of the entire network, to solve the problem of inaccurate assessment. The experiments show that the method improves the accuracy of the assessment of network situation assessment greatly.
{"title":"Network Situation Risk Assessment Based on Vulnerability Correlation Analysis","authors":"X. Nan, R. Chen, Hongtao Tian, Yupeng Liu","doi":"10.1109/PIC53636.2021.9687007","DOIUrl":"https://doi.org/10.1109/PIC53636.2021.9687007","url":null,"abstract":"For the question that situation assessment methods for the analysis of existing vulnerabilities are associated with the lack of analysis of vulnerability assessments, which leads to the poor accuracy assessment, the paper presents a method for network vulnerabilities associated with risk assessment situation analysis. The method improves the existing hierarchical network situation assessment, with the system being divided into three levels, which are loopholes at the bottom, host in the middle, and network system at the top. Based on the security risk indices, we calculate the vulnerability, the host, the entire network system risk index, and evaluate and analyze the security posture of the entire network, to solve the problem of inaccurate assessment. The experiments show that the method improves the accuracy of the assessment of network situation assessment greatly.","PeriodicalId":297239,"journal":{"name":"2021 IEEE International Conference on Progress in Informatics and Computing (PIC)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122143057","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-17DOI: 10.1109/PIC53636.2021.9687051
Bin Zhang, Kenji Isobe, Hun-ok Lim
This paper presents a 1-DOF (degree-of-freedom) power assisting system that can assist elbow flexion motion by using mechanomyogram (MMG) signals. An MMG transducer is attached to the skin of biceps brachii to monitor muscle actions when the arm is moving. To estimate elbow flexion joint torques, a Hill-type muscle model and a musculoskeletal model of the arm are used. The estimated joint torques are substituted into the admittance control system, and the elbow joint angle is calculated. Power assisting experiments are conducted, and the effectiveness of the power assist system is verified.
{"title":"Development of a 1-DOF Elbow Power Assisting System Based on Mechanomyogram Signals","authors":"Bin Zhang, Kenji Isobe, Hun-ok Lim","doi":"10.1109/PIC53636.2021.9687051","DOIUrl":"https://doi.org/10.1109/PIC53636.2021.9687051","url":null,"abstract":"This paper presents a 1-DOF (degree-of-freedom) power assisting system that can assist elbow flexion motion by using mechanomyogram (MMG) signals. An MMG transducer is attached to the skin of biceps brachii to monitor muscle actions when the arm is moving. To estimate elbow flexion joint torques, a Hill-type muscle model and a musculoskeletal model of the arm are used. The estimated joint torques are substituted into the admittance control system, and the elbow joint angle is calculated. Power assisting experiments are conducted, and the effectiveness of the power assist system is verified.","PeriodicalId":297239,"journal":{"name":"2021 IEEE International Conference on Progress in Informatics and Computing (PIC)","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125643386","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-17DOI: 10.1109/PIC53636.2021.9687059
Ziqian Wang, Kaiyu Wang, Jiaru Yang, Zheng Tang
As current mainstream deep learning models based on neural networks have been long criticized because of their complex structures, attempts in formulating a single neural model have raised much attention. Owing to the nonlinear information processing ability, dendritic neuron model (DNM) has shown its great potential in classification problems over the past decades. However, designing an effective learning algorithm for training DNM is still an open question due to the issues of local optima trapping and overfiting caused by traditional back-propagation (BP) algorithm. In this study, a novel functional weight-enhanced differential evolutionary algorithm (termed FWDE) is proposed to solve the aforementioned problems. By introducing Gaussian distribution function into weight generation of fitness-distance balance selection strategy, FWDE obtains significantly better classification accuracy with faster convergence speed compared with other representative non-BP and BP algorithms. The experimental results verify the great performance of FWDE, indicating that DNM with an powerful learning algorithm is considerably more effective.
{"title":"Single Dendritic Neural Classification with Functional Weight-enhanced Differential Evolution","authors":"Ziqian Wang, Kaiyu Wang, Jiaru Yang, Zheng Tang","doi":"10.1109/PIC53636.2021.9687059","DOIUrl":"https://doi.org/10.1109/PIC53636.2021.9687059","url":null,"abstract":"As current mainstream deep learning models based on neural networks have been long criticized because of their complex structures, attempts in formulating a single neural model have raised much attention. Owing to the nonlinear information processing ability, dendritic neuron model (DNM) has shown its great potential in classification problems over the past decades. However, designing an effective learning algorithm for training DNM is still an open question due to the issues of local optima trapping and overfiting caused by traditional back-propagation (BP) algorithm. In this study, a novel functional weight-enhanced differential evolutionary algorithm (termed FWDE) is proposed to solve the aforementioned problems. By introducing Gaussian distribution function into weight generation of fitness-distance balance selection strategy, FWDE obtains significantly better classification accuracy with faster convergence speed compared with other representative non-BP and BP algorithms. The experimental results verify the great performance of FWDE, indicating that DNM with an powerful learning algorithm is considerably more effective.","PeriodicalId":297239,"journal":{"name":"2021 IEEE International Conference on Progress in Informatics and Computing (PIC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125491259","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：