Pub Date : 2021-09-24DOI: 10.1109/ICISCAE52414.2021.9590732
Zhangrong Liu
With the wide popularization of the Internet, advanced technologies such as AI (artificial intelligence) and mega data have also been produced and widely used, effectively promoting the development and progress of related industries. In the era of mega data, people can get valuable information through data mining and analysis. The core goal of building a computer mega data security technology platform under AI environment is to build a corresponding computer mega data security technology platform through AI, so as to realize the protection of user data information. On the basis of simulating human behavior, ML (machine learning) technology enables computers to acquire more new skills and knowledge so as to complete the continuous optimization of organizational structure. This paper introduces the key technologies of network security and expounds the application of ML technology in the field of network security, then analyzes the solutions of ML in network security under the background of mega data, and finally looks forward to the development trend of ML in network security research.
{"title":"Construction of Computer Mega Data Security Technology Platform Based on Machine Learning","authors":"Zhangrong Liu","doi":"10.1109/ICISCAE52414.2021.9590732","DOIUrl":"https://doi.org/10.1109/ICISCAE52414.2021.9590732","url":null,"abstract":"With the wide popularization of the Internet, advanced technologies such as AI (artificial intelligence) and mega data have also been produced and widely used, effectively promoting the development and progress of related industries. In the era of mega data, people can get valuable information through data mining and analysis. The core goal of building a computer mega data security technology platform under AI environment is to build a corresponding computer mega data security technology platform through AI, so as to realize the protection of user data information. On the basis of simulating human behavior, ML (machine learning) technology enables computers to acquire more new skills and knowledge so as to complete the continuous optimization of organizational structure. This paper introduces the key technologies of network security and expounds the application of ML technology in the field of network security, then analyzes the solutions of ML in network security under the background of mega data, and finally looks forward to the development trend of ML in network security research.","PeriodicalId":121049,"journal":{"name":"2021 IEEE 4th International Conference on Information Systems and Computer Aided Education (ICISCAE)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116486207","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 : 2021-09-24DOI: 10.1109/ICISCAE52414.2021.9590741
Zheng Liu, Shanshan Gao, Jing Chi, Huijian Han
In this paper, we propose a novel strategy to accurately evaluate learning status and development potentiality for colleges students based on big data modeling and mining. Firstly, we propose an effective method to calculate students ability achievement based on the big data mining technology via three steps. In step 1, we construct the student ability achievement evaluation system, which reflects the “output-oriented concept”. In step 2, we model the educational big data, which embodies the “continuous improvement concept”. In step 3, we compute students ability achievement based on educational big data mining, which embodies the “student-centered concept”. Secondly, we discuss how to accurately evaluate college students learning status and development potentiality based on the ability achievements. Finally, we conduct a series of experiments to demonstrate the effectiveness of the proposed solution. Experimental results demonstrate that the proposed solution can effectively evaluate learning status and development potentiality, moreover, the proposed solution can recommend suitable courses and career plannings for college students.
{"title":"Research on Accurate Learning Evaluation for College Students Based on Big Data Technology","authors":"Zheng Liu, Shanshan Gao, Jing Chi, Huijian Han","doi":"10.1109/ICISCAE52414.2021.9590741","DOIUrl":"https://doi.org/10.1109/ICISCAE52414.2021.9590741","url":null,"abstract":"In this paper, we propose a novel strategy to accurately evaluate learning status and development potentiality for colleges students based on big data modeling and mining. Firstly, we propose an effective method to calculate students ability achievement based on the big data mining technology via three steps. In step 1, we construct the student ability achievement evaluation system, which reflects the “output-oriented concept”. In step 2, we model the educational big data, which embodies the “continuous improvement concept”. In step 3, we compute students ability achievement based on educational big data mining, which embodies the “student-centered concept”. Secondly, we discuss how to accurately evaluate college students learning status and development potentiality based on the ability achievements. Finally, we conduct a series of experiments to demonstrate the effectiveness of the proposed solution. Experimental results demonstrate that the proposed solution can effectively evaluate learning status and development potentiality, moreover, the proposed solution can recommend suitable courses and career plannings for college students.","PeriodicalId":121049,"journal":{"name":"2021 IEEE 4th International Conference on Information Systems and Computer Aided Education (ICISCAE)","volume":"115 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124159850","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 : 2021-09-24DOI: 10.1109/ICISCAE52414.2021.9590686
J. Duan
With the gradual improvement of the infrastructure of the information society, computer network information security has involved all aspects of people's lives. Therefore, computer network security has also affected the security and economic development of all walks of life when applying computer networks to some extent. Nowadays, all walks of life in society are closely related to computer networks, from state secrets to people's daily entertainment. From the computer's own defects and the related prediction of computer security in the future, it can be known that security is the direction that needs to be focused on at present, and the protective measures and evaluation algorithms for computer network information security need further research. In this paper, from the perspective of computer network information security, based on fuzzy clustering analysis, the computer network security protection problem is evaluated. The computer network information security protection strategy and evaluation algorithm introduced in this paper are of great significance for improving the computer network security level.
{"title":"Research on Computer Network Information Security Protection Strategy and Evaluation Algorithm Based on Fuzzy Clustering","authors":"J. Duan","doi":"10.1109/ICISCAE52414.2021.9590686","DOIUrl":"https://doi.org/10.1109/ICISCAE52414.2021.9590686","url":null,"abstract":"With the gradual improvement of the infrastructure of the information society, computer network information security has involved all aspects of people's lives. Therefore, computer network security has also affected the security and economic development of all walks of life when applying computer networks to some extent. Nowadays, all walks of life in society are closely related to computer networks, from state secrets to people's daily entertainment. From the computer's own defects and the related prediction of computer security in the future, it can be known that security is the direction that needs to be focused on at present, and the protective measures and evaluation algorithms for computer network information security need further research. In this paper, from the perspective of computer network information security, based on fuzzy clustering analysis, the computer network security protection problem is evaluated. The computer network information security protection strategy and evaluation algorithm introduced in this paper are of great significance for improving the computer network security level.","PeriodicalId":121049,"journal":{"name":"2021 IEEE 4th International Conference on Information Systems and Computer Aided Education (ICISCAE)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127606656","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 : 2021-09-24DOI: 10.1109/ICISCAE52414.2021.9590706
Xiaoyu Wang, Xue-Qin Lu, Zhonghua Liang, Gang Liu
To solve the problem of low utilization of underlying network resources, this paper proposes a network resource backup algorithm based on a multi-attribute analytic hierarchy process in IPv6 Networks. In the steps of calculating and ranking the importance of the underlying network nodes and underlying links, analysis is performed from the three dimensions of network characteristics, resource utilization, and power business importance of each underlying network node and underlying network link. To analyze the weight function of multiple attributes in the importance of resources more fairly, the attributes of the underlying network nodes and underlying network links are analyzed based on the multi-attribute hierarchical analysis algorithm. When performing resource backups on the underlying network nodes and underlying network links, it is evaluated from the two dimensions of whether the resource utilization rate exceeds the threshold and the importance of the resources, to improve the rationality of the underlying network resource backup mechanism. In the experimental part, it is verified that the algorithm in this paper has obtained good results in the two dimensions of the virtual network mapping success rate and the average utilization rate of the underlying network resources.
{"title":"Network Resource Backup Algorithm Based on Multi-Attribute Analytic Hierarchy Process in IPv6 Networks","authors":"Xiaoyu Wang, Xue-Qin Lu, Zhonghua Liang, Gang Liu","doi":"10.1109/ICISCAE52414.2021.9590706","DOIUrl":"https://doi.org/10.1109/ICISCAE52414.2021.9590706","url":null,"abstract":"To solve the problem of low utilization of underlying network resources, this paper proposes a network resource backup algorithm based on a multi-attribute analytic hierarchy process in IPv6 Networks. In the steps of calculating and ranking the importance of the underlying network nodes and underlying links, analysis is performed from the three dimensions of network characteristics, resource utilization, and power business importance of each underlying network node and underlying network link. To analyze the weight function of multiple attributes in the importance of resources more fairly, the attributes of the underlying network nodes and underlying network links are analyzed based on the multi-attribute hierarchical analysis algorithm. When performing resource backups on the underlying network nodes and underlying network links, it is evaluated from the two dimensions of whether the resource utilization rate exceeds the threshold and the importance of the resources, to improve the rationality of the underlying network resource backup mechanism. In the experimental part, it is verified that the algorithm in this paper has obtained good results in the two dimensions of the virtual network mapping success rate and the average utilization rate of the underlying network resources.","PeriodicalId":121049,"journal":{"name":"2021 IEEE 4th International Conference on Information Systems and Computer Aided Education (ICISCAE)","volume":"93 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127227686","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 : 2021-09-24DOI: 10.1109/ICISCAE52414.2021.9590766
Kaibin Xu, Junpei Zhong, Kristiina Jokinen
Based on previous studies on the co-occurrence between the types of laughter and conversation contents, this paper uses the attention mechanism to further study the types of laughter and the specific wording in the conversation. We assume that the types of laughter also co-occurs with specific words, as well as a certain pause, in the conversation. We have established a neural network learning model based on the attention mechanism. The results of our model show that certain words in the conversation are related to specific types of laughter.
{"title":"It is Time to Laugh: Discovering Specific Contexts for Laughter with Attention Mechanism","authors":"Kaibin Xu, Junpei Zhong, Kristiina Jokinen","doi":"10.1109/ICISCAE52414.2021.9590766","DOIUrl":"https://doi.org/10.1109/ICISCAE52414.2021.9590766","url":null,"abstract":"Based on previous studies on the co-occurrence between the types of laughter and conversation contents, this paper uses the attention mechanism to further study the types of laughter and the specific wording in the conversation. We assume that the types of laughter also co-occurs with specific words, as well as a certain pause, in the conversation. We have established a neural network learning model based on the attention mechanism. The results of our model show that certain words in the conversation are related to specific types of laughter.","PeriodicalId":121049,"journal":{"name":"2021 IEEE 4th International Conference on Information Systems and Computer Aided Education (ICISCAE)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131319416","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 : 2021-09-24DOI: 10.1109/ICISCAE52414.2021.9590798
Zhou Bin
In order to solve the problem of information resource sharing caused by information isolated island, this paper proposes a cross system online information resource sharing platform model based on the existing information system of colleges and universities, which is role and transaction oriented. The online resource sharing platform based on this model establishes a specific structure of intermediate database on the basis of the existing isolated information system, adopts the strategy of non authoritative data accurate association matching authoritative data, realizes its interconnection with the existing system databases through specific program algorithm logic, constructs a distributed interconnected database, and realizes cross database data association matching. After solving the problem of information island fundamentally, an information resource sharing platform that can define any role or transaction can be built to meet the current cross system information resource sharing needs. Based on the existing system management information, the purpose of accurately sharing information across systems by roles or transactions through the new platform is realized. Through the implementation and practical application of the model technology, it is proved that it can better solve the problem of information resource sharing caused by information island, and realize the accurate cross system information resource sharing service.
{"title":"A model of information sharing based on isolated information system","authors":"Zhou Bin","doi":"10.1109/ICISCAE52414.2021.9590798","DOIUrl":"https://doi.org/10.1109/ICISCAE52414.2021.9590798","url":null,"abstract":"In order to solve the problem of information resource sharing caused by information isolated island, this paper proposes a cross system online information resource sharing platform model based on the existing information system of colleges and universities, which is role and transaction oriented. The online resource sharing platform based on this model establishes a specific structure of intermediate database on the basis of the existing isolated information system, adopts the strategy of non authoritative data accurate association matching authoritative data, realizes its interconnection with the existing system databases through specific program algorithm logic, constructs a distributed interconnected database, and realizes cross database data association matching. After solving the problem of information island fundamentally, an information resource sharing platform that can define any role or transaction can be built to meet the current cross system information resource sharing needs. Based on the existing system management information, the purpose of accurately sharing information across systems by roles or transactions through the new platform is realized. Through the implementation and practical application of the model technology, it is proved that it can better solve the problem of information resource sharing caused by information island, and realize the accurate cross system information resource sharing service.","PeriodicalId":121049,"journal":{"name":"2021 IEEE 4th International Conference on Information Systems and Computer Aided Education (ICISCAE)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132430805","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 : 2021-09-24DOI: 10.1109/ICISCAE52414.2021.9590780
Yinquan Liu, Zhuang Chen
With the development of neural style transfer and generative adversarial network, the research of text effect style transfer has appeared. The text effect style transfer aims to render text images with style images to produce text effects images. However, for more complex text, the existing methods will generate unrecognizable font images. Therefore, we propose to add morphological methods to the glyph transformation to limit the degree of glyph transformation, and add distance transformation loss when training the texture network to limit the texture transfer, so as to improve the overall transformation effect. Experiments show that, compared with other existing technologies, our proposed method is more suitable for stylizing complex glyph images.
{"title":"Research on GAN-based Text Effects Style Transfer","authors":"Yinquan Liu, Zhuang Chen","doi":"10.1109/ICISCAE52414.2021.9590780","DOIUrl":"https://doi.org/10.1109/ICISCAE52414.2021.9590780","url":null,"abstract":"With the development of neural style transfer and generative adversarial network, the research of text effect style transfer has appeared. The text effect style transfer aims to render text images with style images to produce text effects images. However, for more complex text, the existing methods will generate unrecognizable font images. Therefore, we propose to add morphological methods to the glyph transformation to limit the degree of glyph transformation, and add distance transformation loss when training the texture network to limit the texture transfer, so as to improve the overall transformation effect. Experiments show that, compared with other existing technologies, our proposed method is more suitable for stylizing complex glyph images.","PeriodicalId":121049,"journal":{"name":"2021 IEEE 4th International Conference on Information Systems and Computer Aided Education (ICISCAE)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130156423","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 : 2021-09-24DOI: 10.1109/ICISCAE52414.2021.9590778
Xuhan Wu, Wentao Wang, Hong Wang
In order to reasonably describe the uncertainty of output of instruments and actuators, a concept of transient stability control based on online pre-decision of mapping cloud is proposed. This paper describes an online pre-decision transient stability control system developed by Electric Power Automation Research Institute, which integrates quantitative analysis technology of motion stability of nonautonomous nonlinear multi-rigid-body system, object-oriented distributed parallel processing platform and regional stability control device. The system online tracks the system working conditions using the stability margin index of the controlled power system provided by EEAC, and quickly searches for the control strategy that corresponds to a given fault set, satisfies a certain system stability margin value, and has the least control cost, so as to correct The development and change of the power grid and the change of operation mode have strong adaptive ability. The system is an organic combination of transient stability quantitative analysis technology EEAC and mature regional stability control device FWK, and is a new product of domestic safety and stability control technology.
{"title":"Transient stability control system based on mapping cloud and online pre decision","authors":"Xuhan Wu, Wentao Wang, Hong Wang","doi":"10.1109/ICISCAE52414.2021.9590778","DOIUrl":"https://doi.org/10.1109/ICISCAE52414.2021.9590778","url":null,"abstract":"In order to reasonably describe the uncertainty of output of instruments and actuators, a concept of transient stability control based on online pre-decision of mapping cloud is proposed. This paper describes an online pre-decision transient stability control system developed by Electric Power Automation Research Institute, which integrates quantitative analysis technology of motion stability of nonautonomous nonlinear multi-rigid-body system, object-oriented distributed parallel processing platform and regional stability control device. The system online tracks the system working conditions using the stability margin index of the controlled power system provided by EEAC, and quickly searches for the control strategy that corresponds to a given fault set, satisfies a certain system stability margin value, and has the least control cost, so as to correct The development and change of the power grid and the change of operation mode have strong adaptive ability. The system is an organic combination of transient stability quantitative analysis technology EEAC and mature regional stability control device FWK, and is a new product of domestic safety and stability control technology.","PeriodicalId":121049,"journal":{"name":"2021 IEEE 4th International Conference on Information Systems and Computer Aided Education (ICISCAE)","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114235566","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 : 2021-09-24DOI: 10.1109/ICISCAE52414.2021.9590802
Jianwei Li, Zhibing Zhang, Kai-Zhong Zhou, Dan Luo
In a IPv6 network environment, how to restore as many virtual network services as possible under the constraints of limited resources is an urgent problem to be solved. To solve this problem, this paper models the virtual network service recovery problem as a problem of maximizing the number of failed services recovery. A particle swarm-based virtual service failure recovery algorithm in IPv6 networks is proposed. The algorithm first constructs the faulty resource and virtual service as a two-layer correlation model. Secondly, the network resource recovery problem is modeled as a particle swarm problem, and the particle swarm optimization algorithm is used to solve it. In the experimental part, by comparing with existing algorithms, it is verified that the algorithm in this paper can recover more virtual network services that have failed.
{"title":"Virtual Service Failure Recovery Algorithm Based on Particle Swarm in IPv6 Networks","authors":"Jianwei Li, Zhibing Zhang, Kai-Zhong Zhou, Dan Luo","doi":"10.1109/ICISCAE52414.2021.9590802","DOIUrl":"https://doi.org/10.1109/ICISCAE52414.2021.9590802","url":null,"abstract":"In a IPv6 network environment, how to restore as many virtual network services as possible under the constraints of limited resources is an urgent problem to be solved. To solve this problem, this paper models the virtual network service recovery problem as a problem of maximizing the number of failed services recovery. A particle swarm-based virtual service failure recovery algorithm in IPv6 networks is proposed. The algorithm first constructs the faulty resource and virtual service as a two-layer correlation model. Secondly, the network resource recovery problem is modeled as a particle swarm problem, and the particle swarm optimization algorithm is used to solve it. In the experimental part, by comparing with existing algorithms, it is verified that the algorithm in this paper can recover more virtual network services that have failed.","PeriodicalId":121049,"journal":{"name":"2021 IEEE 4th International Conference on Information Systems and Computer Aided Education (ICISCAE)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114721023","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 : 2021-09-24DOI: 10.1109/ICISCAE52414.2021.9590701
Bo Qin, Wei He, Bin Zhang, Jingchen Li
As a significant issue in the development of Smart City, traffic light control is closely related to public transport. Multi-agent reinforcement learning is a feasible paradigm to control decentralized systems. In this work, we design a recurrent communication module for traffic light control systems. Aiming at lightweight decentralized learning, we decouple the interaction among entities, using an improved recurrent neural network to share messages. Regarding traffic lights as a partially observable Markov Decision Process, we develop a complete reinforcement learning model. Every light receives messages from its neighbors through the improved recurrent neural network. For the further lights, its neighbor lights can play the role of intermediation to transmit messages. By our communication framework, traffic lights can learn their control policies in a decentralized way, and distribution control can be executed through a communication channel. Several simulated experiments show that our method outperforms the fixed time method and the existing reinforcement learning-based frameworks under different traffic demands.
{"title":"A Multi-Agent Reinforcement Learning Framework with Recurrent Communication Module for Traffic Light Control","authors":"Bo Qin, Wei He, Bin Zhang, Jingchen Li","doi":"10.1109/ICISCAE52414.2021.9590701","DOIUrl":"https://doi.org/10.1109/ICISCAE52414.2021.9590701","url":null,"abstract":"As a significant issue in the development of Smart City, traffic light control is closely related to public transport. Multi-agent reinforcement learning is a feasible paradigm to control decentralized systems. In this work, we design a recurrent communication module for traffic light control systems. Aiming at lightweight decentralized learning, we decouple the interaction among entities, using an improved recurrent neural network to share messages. Regarding traffic lights as a partially observable Markov Decision Process, we develop a complete reinforcement learning model. Every light receives messages from its neighbors through the improved recurrent neural network. For the further lights, its neighbor lights can play the role of intermediation to transmit messages. By our communication framework, traffic lights can learn their control policies in a decentralized way, and distribution control can be executed through a communication channel. Several simulated experiments show that our method outperforms the fixed time method and the existing reinforcement learning-based frameworks under different traffic demands.","PeriodicalId":121049,"journal":{"name":"2021 IEEE 4th International Conference on Information Systems and Computer Aided Education (ICISCAE)","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128378325","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：