Pub Date : 2023-04-01DOI: 10.53106/199115992023043402011
Mingxin Yang Mingxin Yang, Lei Feng Mingxin Yang
As the information technology develops, network attacks have become complex and diverse. To improve the effectiveness and accuracy of data security defense strategies, an optimal defense method based on improving evolutionary game model between heterogeneous groups is proposed. Specifically, based on traditional evolutionary game theory, the player type space is added to divide the heterogeneous groups, and the group type and game strategy are extended to N to solve the problems in heterogeneous groups. Considering the game is interfered by the environment, a set of dynamic environment functions is added to increase the adaptability of the model when dealing with changing complex networks. Taking into account the influence of information communication within the group, the information flow degree is added to increase the accuracy of evolution rate and solve the problem that traditional model cannot reveal the difference in the evolution rate of players. Based on the new model, taking the game between two types of invaders and one type of defender as an example, the calculation method of evolution direction at any time and the judgment method for the stability of equilibrium point are discussed. Finally, the effectiveness of the improved model is verified through the comparison of simulation experiments, and a new scheme is provided for current network data protection.
{"title":"Optimal Defense Strategy for Data Security Based on Improving Evolutionary Game Model between Heterogeneous Groups","authors":"Mingxin Yang Mingxin Yang, Lei Feng Mingxin Yang","doi":"10.53106/199115992023043402011","DOIUrl":"https://doi.org/10.53106/199115992023043402011","url":null,"abstract":"\u0000 As the information technology develops, network attacks have become complex and diverse. To improve the effectiveness and accuracy of data security defense strategies, an optimal defense method based on improving evolutionary game model between heterogeneous groups is proposed. Specifically, based on traditional evolutionary game theory, the player type space is added to divide the heterogeneous groups, and the group type and game strategy are extended to N to solve the problems in heterogeneous groups. Considering the game is interfered by the environment, a set of dynamic environment functions is added to increase the adaptability of the model when dealing with changing complex networks. Taking into account the influence of information communication within the group, the information flow degree is added to increase the accuracy of evolution rate and solve the problem that traditional model cannot reveal the difference in the evolution rate of players. Based on the new model, taking the game between two types of invaders and one type of defender as an example, the calculation method of evolution direction at any time and the judgment method for the stability of equilibrium point are discussed. Finally, the effectiveness of the improved model is verified through the comparison of simulation experiments, and a new scheme is provided for current network data protection.\u0000 \u0000","PeriodicalId":345067,"journal":{"name":"電腦學刊","volume":"50 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126673561","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 : 2023-04-01DOI: 10.53106/199115992023043402008
Haiyan Wang Haiyan Wang, Haibing Mu Haiyan Wang
Internet of Vehicles (IoV) plays an important role in enhancing the intelligence of social transportation services. However, there are existing such as privacy leakage, computational complexity and low efficiency on V2R authentication protocols. To solve these problems, a lightweight V2R authentication protocol according to physical unclonable function (PUF) and Chebyshev chaotic map is proposed. The lightweight property of PUF in this scheme can solve the resource constraint problem of the On-Board Unit (OBU) effectively. The fuzzy extractor can correct for small variations in PUF response and improve the accuracy of data transmission. Besides, Chebyshev chaotic map with good cryptographic properties establishes a secure session key while achieving mutual authentication of V2R. Finally, simulation results show that the scheme combining PUF, chaotic map, and fuzzy extractor in this paper saves 4.7% to 49% in communication and calculation overhead comparing with existing protocols. In terms of security, our scheme can also meet the requirements well in the V2R authentication protocol for IoV.
{"title":"A Lightweight V2R Authentication Protocol Based on PUF and Chebyshev Chaotic Map","authors":"Haiyan Wang Haiyan Wang, Haibing Mu Haiyan Wang","doi":"10.53106/199115992023043402008","DOIUrl":"https://doi.org/10.53106/199115992023043402008","url":null,"abstract":"\u0000 Internet of Vehicles (IoV) plays an important role in enhancing the intelligence of social transportation services. However, there are existing such as privacy leakage, computational complexity and low efficiency on V2R authentication protocols. To solve these problems, a lightweight V2R authentication protocol according to physical unclonable function (PUF) and Chebyshev chaotic map is proposed. The lightweight property of PUF in this scheme can solve the resource constraint problem of the On-Board Unit (OBU) effectively. The fuzzy extractor can correct for small variations in PUF response and improve the accuracy of data transmission. Besides, Chebyshev chaotic map with good cryptographic properties establishes a secure session key while achieving mutual authentication of V2R. Finally, simulation results show that the scheme combining PUF, chaotic map, and fuzzy extractor in this paper saves 4.7% to 49% in communication and calculation overhead comparing with existing protocols. In terms of security, our scheme can also meet the requirements well in the V2R authentication protocol for IoV.\u0000 \u0000","PeriodicalId":345067,"journal":{"name":"電腦學刊","volume":"173 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117349391","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}
According to the relevant requirements of the Code for Digital Campus of Vocational Colleges of the Ministry of Education, combined with the characteristics of teaching and education of vocational colleges and the needs of information construction, this paper analyzes the new requirements and characteristics of digital campus construction of vocational colleges, and puts forward the connotation and construction principles of digital campus. On this basis, the overall construction framework and construction contents of digital campus of vocational colleges are given. Finally, according to the characteristics of vocational college informatization construction, this paper puts forward some suggestions to promote the implementation of vocational college digital campus construction, which can provide reference for promoting the modernization of vocational education with the help of informatization.
{"title":"Research on The Construction of Digital Campus for Vocational Colleges","authors":"Yongjun Wei Yongjun Wei, Qiumi Qin Yongjun Wei, Jingling Xiao Qiumi Qin, Jun Yin Jingling Xiao, Wufeng Chen Jun Yin, Guangfa Liang Wufeng Chen","doi":"10.53106/199115992023043402017","DOIUrl":"https://doi.org/10.53106/199115992023043402017","url":null,"abstract":"\u0000 According to the relevant requirements of the Code for Digital Campus of Vocational Colleges of the Ministry of Education, combined with the characteristics of teaching and education of vocational colleges and the needs of information construction, this paper analyzes the new requirements and characteristics of digital campus construction of vocational colleges, and puts forward the connotation and construction principles of digital campus. On this basis, the overall construction framework and construction contents of digital campus of vocational colleges are given. Finally, according to the characteristics of vocational college informatization construction, this paper puts forward some suggestions to promote the implementation of vocational college digital campus construction, which can provide reference for promoting the modernization of vocational education with the help of informatization.\u0000 \u0000","PeriodicalId":345067,"journal":{"name":"電腦學刊","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129050735","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 : 2023-04-01DOI: 10.53106/199115992023043402001
Guo-Long Yu Guo-Long Yu, Zhong-Wei Cui Guo-Long Yu, Qiong-Fang Yuan Zhong-Wei Cui
In image segmentation, FCM clustering algorithm can not find the optimal initial clustering center and fall into local extremum, which leads to the decrease of image segmentation accuracy. The PSO algorithm has strong optimization ability, so a new method based on improved PSO algorithm is proposed to optimize the FCM clustering center selection. Firstly, the optimization performance of the PSO algorithm is improved. The distance difference between each particle and the optimal particle is calculated, and the maximum distance difference is selected. The ratio of the distance difference to the maximum distance difference and the aggregation degree of particles are used to construct the natural exponential function. This natural exponential function is used to improve the calculation method of inertia weight value of PSO algorithm, so that the farther the particle is away from the optimal position, the larger the inertia weight value it will get, the stronger the global search ability of particle; on the contrary, the smaller the inertia weight value, the stronger the local search ability of particle, so as to improve the optimization ability of PSO algorithm. The improved PSO algorithm is called DDPSO (Distance Difference PSO). Then the optimized FCM algorithm is applied to the segmentation of standard image and eggshell damaged image to improve the accuracy of image segmentation. Finally, the experimental results show that the FCM algorithm optimized by DDPSO has higher segmentation accuracy than the traditional method.
{"title":"Image Segmentation Method Based on Improved PSO Optimized FCM Algorithm and Its Application","authors":"Guo-Long Yu Guo-Long Yu, Zhong-Wei Cui Guo-Long Yu, Qiong-Fang Yuan Zhong-Wei Cui","doi":"10.53106/199115992023043402001","DOIUrl":"https://doi.org/10.53106/199115992023043402001","url":null,"abstract":"\u0000 In image segmentation, FCM clustering algorithm can not find the optimal initial clustering center and fall into local extremum, which leads to the decrease of image segmentation accuracy. The PSO algorithm has strong optimization ability, so a new method based on improved PSO algorithm is proposed to optimize the FCM clustering center selection. Firstly, the optimization performance of the PSO algorithm is improved. The distance difference between each particle and the optimal particle is calculated, and the maximum distance difference is selected. The ratio of the distance difference to the maximum distance difference and the aggregation degree of particles are used to construct the natural exponential function. This natural exponential function is used to improve the calculation method of inertia weight value of PSO algorithm, so that the farther the particle is away from the optimal position, the larger the inertia weight value it will get, the stronger the global search ability of particle; on the contrary, the smaller the inertia weight value, the stronger the local search ability of particle, so as to improve the optimization ability of PSO algorithm. The improved PSO algorithm is called DDPSO (Distance Difference PSO). Then the optimized FCM algorithm is applied to the segmentation of standard image and eggshell damaged image to improve the accuracy of image segmentation. Finally, the experimental results show that the FCM algorithm optimized by DDPSO has higher segmentation accuracy than the traditional method.\u0000 \u0000","PeriodicalId":345067,"journal":{"name":"電腦學刊","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132096413","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 : 2023-04-01DOI: 10.53106/199115992023043402015
Jian Wang Jian Wang, Dong-Liang Fan Jian Wang, Jin-Ping Du Dong-Liang Fan, Lei Geng Jin-Ping Du, Ya-Jin Hou Lei Geng
Lithium batteries are widely used in new energy vehicles and electronic equipment. Aiming at the typical defects that are easy to occur in the production process of lithium batteries, this paper improves the performance and recognition accuracy of the algorithm by integrating void convolution and attention mechanism into the YOLOv5 basic framework. At the same time, whale algorithm is used to automatically optimize the algorithm parameters in the process of optimization. Finally, through simulation experiments. This method realizes the rapid and accurate identification of lithium battery defects in the rapid production process of automatic production line.
{"title":"Research on Artificial Intelligence Detection Method of Lithium Battery Surface Defects for Production Line","authors":"Jian Wang Jian Wang, Dong-Liang Fan Jian Wang, Jin-Ping Du Dong-Liang Fan, Lei Geng Jin-Ping Du, Ya-Jin Hou Lei Geng","doi":"10.53106/199115992023043402015","DOIUrl":"https://doi.org/10.53106/199115992023043402015","url":null,"abstract":"\u0000 Lithium batteries are widely used in new energy vehicles and electronic equipment. Aiming at the typical defects that are easy to occur in the production process of lithium batteries, this paper improves the performance and recognition accuracy of the algorithm by integrating void convolution and attention mechanism into the YOLOv5 basic framework. At the same time, whale algorithm is used to automatically optimize the algorithm parameters in the process of optimization. Finally, through simulation experiments. This method realizes the rapid and accurate identification of lithium battery defects in the rapid production process of automatic production line.\u0000 \u0000","PeriodicalId":345067,"journal":{"name":"電腦學刊","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131568860","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 : 2023-04-01DOI: 10.53106/199115992023043402018
Yi-Hui Chen Yi-Hui Chen
Axial hydraulic piston pump is widely used in industrial production due to its high pressure resistance and large displacement characteristics, but high pressure and large displacement are also the main causes of piston pump failure. Starting from the fault mechanism of the axial hydraulic piston pump, this paper analyzes and studies the signal characteristics of the fault, and establishes the fault signal acquisition and analysis model. Finally, it discusses the construction of the diagnosis system from both hardware and software, so that the processed typical fault signals can be sent into the intelligent diagnosis system to determine the fault type. Finally, the method in this paper is verified by experiments, which proves the reliability and effectiveness of the diagnosis system.
{"title":"A Computer-Aided Intelligent Fault Diagnosis Method for Axial Hydraulic Piston Pump","authors":"Yi-Hui Chen Yi-Hui Chen","doi":"10.53106/199115992023043402018","DOIUrl":"https://doi.org/10.53106/199115992023043402018","url":null,"abstract":"\u0000 Axial hydraulic piston pump is widely used in industrial production due to its high pressure resistance and large displacement characteristics, but high pressure and large displacement are also the main causes of piston pump failure. Starting from the fault mechanism of the axial hydraulic piston pump, this paper analyzes and studies the signal characteristics of the fault, and establishes the fault signal acquisition and analysis model. Finally, it discusses the construction of the diagnosis system from both hardware and software, so that the processed typical fault signals can be sent into the intelligent diagnosis system to determine the fault type. Finally, the method in this paper is verified by experiments, which proves the reliability and effectiveness of the diagnosis system.\u0000 \u0000","PeriodicalId":345067,"journal":{"name":"電腦學刊","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124801733","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 : 2023-04-01DOI: 10.53106/199115992023043402006
Chen Wang Chen Wang, Bingchun Liu Chen Wang, Jiali Chen Bingchun Liu, Xiaogang Yu Jiali Chen
Air pollution has become one of the important challenges restricting the sustainable development of cities. Therefore, it is of great significance to achieve accurate prediction of Air Quality Index (AQI). Long Short Term Memory (LSTM) is a deep learning method suitable for learning time series data. Considering its superiority in processing time series data, this study established an LSTM forecasting model suitable for air quality index forecasting. First, we focus on optimizing the feature metrics of the model input through Information Gain (IG). Second, the prediction results of the LSTM model are compared with other machine learning models. At the same time the time step aspect of the LSTM model is used with selective experiments to ensure that model validation works properly. The results show that compared with other machine learning models, the LSTM model constructed in this paper is more suitable for the prediction of air quality index.
大气污染已成为制约城市可持续发展的重要挑战之一。因此，实现空气质量指数(AQI)的准确预测具有重要意义。长短期记忆(LSTM)是一种适合学习时间序列数据的深度学习方法。考虑到LSTM在处理时间序列数据方面的优势，本研究建立了适合于空气质量指数预测的LSTM预测模型。首先，我们专注于通过信息增益(Information Gain, IG)优化模型输入的特征度量。其次，将LSTM模型的预测结果与其他机器学习模型进行比较。同时对LSTM模型的时间步长方面进行了选择性实验，以确保模型验证工作正常进行。结果表明，与其他机器学习模型相比，本文构建的LSTM模型更适合于空气质量指数的预测。
{"title":"Air Quality Index Prediction Based on a Long Short-Term Memory Artificial Neural Network Model","authors":"Chen Wang Chen Wang, Bingchun Liu Chen Wang, Jiali Chen Bingchun Liu, Xiaogang Yu Jiali Chen","doi":"10.53106/199115992023043402006","DOIUrl":"https://doi.org/10.53106/199115992023043402006","url":null,"abstract":"\u0000 Air pollution has become one of the important challenges restricting the sustainable development of cities. Therefore, it is of great significance to achieve accurate prediction of Air Quality Index (AQI). Long Short Term Memory (LSTM) is a deep learning method suitable for learning time series data. Considering its superiority in processing time series data, this study established an LSTM forecasting model suitable for air quality index forecasting. First, we focus on optimizing the feature metrics of the model input through Information Gain (IG). Second, the prediction results of the LSTM model are compared with other machine learning models. At the same time the time step aspect of the LSTM model is used with selective experiments to ensure that model validation works properly. The results show that compared with other machine learning models, the LSTM model constructed in this paper is more suitable for the prediction of air quality index.\u0000 \u0000","PeriodicalId":345067,"journal":{"name":"電腦學刊","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129067375","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 : 2023-04-01DOI: 10.53106/199115992023043402013
Hao-Qiang Xu Hao-Qiang Xu, Jian-Dong Liu Hao-Qiang Xu
The traditional video encryption algorithm only encrypts video images, which has the problems of an extended time-consuming algorithm and poor format retention. To improve the efficiency of video encryption, this paper proposes a multi-link selective video encryption algorithm based on the Cross Coupled Map Lattices system by combining H.264/AVC video coding structure. The algorithm reduces the amount of encrypted data while ensuring encryption security to satisfy the needs of video encryption security and real-time performance. The encryption algorithm’s security and visual encryption effect are analyzed subjectively and objectively. The experimental results show that the encryption scheme has an excellent visual encryption effect and strong attack resistance, the encryption time consumption is low, and the video format remains unchanged. It can be applied to real-time video encryption occasions such as video conferences.
{"title":"Research on Video Encryption Technology Based on Cross Coupled Map Lattices System","authors":"Hao-Qiang Xu Hao-Qiang Xu, Jian-Dong Liu Hao-Qiang Xu","doi":"10.53106/199115992023043402013","DOIUrl":"https://doi.org/10.53106/199115992023043402013","url":null,"abstract":"\u0000 The traditional video encryption algorithm only encrypts video images, which has the problems of an extended time-consuming algorithm and poor format retention. To improve the efficiency of video encryption, this paper proposes a multi-link selective video encryption algorithm based on the Cross Coupled Map Lattices system by combining H.264/AVC video coding structure. The algorithm reduces the amount of encrypted data while ensuring encryption security to satisfy the needs of video encryption security and real-time performance. The encryption algorithm’s security and visual encryption effect are analyzed subjectively and objectively. The experimental results show that the encryption scheme has an excellent visual encryption effect and strong attack resistance, the encryption time consumption is low, and the video format remains unchanged. It can be applied to real-time video encryption occasions such as video conferences.\u0000 \u0000","PeriodicalId":345067,"journal":{"name":"電腦學刊","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130037567","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}
With the development of automobile technology, intelligent vehicle and automatic driving technology will make due contributions to reducing traffic accidents. This paper aims to improve the dynamic identification and tracking technology in the current intelligent vehicle and automatic driving. First, it is improved based on the MobileNet V2 backbone network, and then a new tracking model framework is designed combining with the SiamRPN single target tracker. Secondly, it integrates space-time tracking clues to improve the stability and robustness of the algorithm. Finally, it constructs a pedestrian dynamic identification algorithm based on the dynamic pedestrian factors in the driving process. Through the training of data sets and video tracking experiments, the performance of the algorithm in this paper is proved quantitatively and qualitatively.
{"title":"Research on Dynamic Recognition and Tracking Technology for Complex Scenes of Automatic Driving","authors":"Shuai-Wu Zhang Shuai-Wu Zhang, Yu-Mei Zhao Shuai-Wu Zhang, Xiang-Lian Yang Yu-Mei Zhao","doi":"10.53106/199115992023043402016","DOIUrl":"https://doi.org/10.53106/199115992023043402016","url":null,"abstract":"\u0000 With the development of automobile technology, intelligent vehicle and automatic driving technology will make due contributions to reducing traffic accidents. This paper aims to improve the dynamic identification and tracking technology in the current intelligent vehicle and automatic driving. First, it is improved based on the MobileNet V2 backbone network, and then a new tracking model framework is designed combining with the SiamRPN single target tracker. Secondly, it integrates space-time tracking clues to improve the stability and robustness of the algorithm. Finally, it constructs a pedestrian dynamic identification algorithm based on the dynamic pedestrian factors in the driving process. Through the training of data sets and video tracking experiments, the performance of the algorithm in this paper is proved quantitatively and qualitatively.\u0000 \u0000","PeriodicalId":345067,"journal":{"name":"電腦學刊","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130061473","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 : 2023-04-01DOI: 10.53106/199115992023043402004
Yan Kong Yan Kong, Yefeng Rui Yan Kong, Chih-Hsien Hsia Yefeng Rui
Recent years have witnessed the big success deep reinforcement learning achieved in the domain of card and board games, such as Go, chess and Texas Hold’em poker. However, Dou Di Zhu, a traditional Chinese card game, is still a challenging task for deep reinforcement learning methods due to the enormous action space and the sparse and delayed reward of each action from the environment. Basic reinforcement learning algorithms are more effective in the simple environments which have small action spaces and valuable and concrete reward functions, and unfortunately, are shown not be able to deal with Dou Di Zhu satisfactorily. This work introduces an approach named Two-steps Q-Network based on DQN to playing Dou Di Zhu, which compresses the huge action space through dividing it into two parts according to the rules of Dou Di Zhu and fills in the sparse rewards using inverse reinforcement learning (IRL) through abstracting the reward function from experts’ demonstrations. It is illustrated by the experiments that two-steps Q-network gains great advancements compared with DQN used in Dou Di Zhu.
{"title":"A Deep Reinforcement Learning-Based Approach in Porker Game","authors":"Yan Kong Yan Kong, Yefeng Rui Yan Kong, Chih-Hsien Hsia Yefeng Rui","doi":"10.53106/199115992023043402004","DOIUrl":"https://doi.org/10.53106/199115992023043402004","url":null,"abstract":"\u0000 Recent years have witnessed the big success deep reinforcement learning achieved in the domain of card and board games, such as Go, chess and Texas Hold’em poker. However, Dou Di Zhu, a traditional Chinese card game, is still a challenging task for deep reinforcement learning methods due to the enormous action space and the sparse and delayed reward of each action from the environment. Basic reinforcement learning algorithms are more effective in the simple environments which have small action spaces and valuable and concrete reward functions, and unfortunately, are shown not be able to deal with Dou Di Zhu satisfactorily. This work introduces an approach named Two-steps Q-Network based on DQN to playing Dou Di Zhu, which compresses the huge action space through dividing it into two parts according to the rules of Dou Di Zhu and fills in the sparse rewards using inverse reinforcement learning (IRL) through abstracting the reward function from experts’ demonstrations. It is illustrated by the experiments that two-steps Q-network gains great advancements compared with DQN used in Dou Di Zhu.\u0000 \u0000","PeriodicalId":345067,"journal":{"name":"電腦學刊","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133314014","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：