Pub Date : 1999-07-10DOI: 10.1109/IJCNN.1999.830745
Jie Si, Rilun Tang
In this paper, three types of chess endgames were studied and three layer feedforward neural networks were applied to learn the hidden rules in chess endgames. The purpose of this paper is to convert the symbolic rules of chess endgames into numerical information that neural networks can learn. The neural networks have been proved efficient in learning and playing some simple cases of chess endgames.
{"title":"Trained neural networks play chess endgames","authors":"Jie Si, Rilun Tang","doi":"10.1109/IJCNN.1999.830745","DOIUrl":"https://doi.org/10.1109/IJCNN.1999.830745","url":null,"abstract":"In this paper, three types of chess endgames were studied and three layer feedforward neural networks were applied to learn the hidden rules in chess endgames. The purpose of this paper is to convert the symbolic rules of chess endgames into numerical information that neural networks can learn. The neural networks have been proved efficient in learning and playing some simple cases of chess endgames.","PeriodicalId":157719,"journal":{"name":"IJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No.99CH36339)","volume":"99 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116808441","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 : 1999-07-10DOI: 10.1109/IJCNN.1999.836219
R. Bharadwaj, A. Parlos, H. Toliyat, S. Menon
Effective sensorless speed estimation is desirable for both online condition monitoring of induction motor and sensorless adjustable speed AC drive applications. In this paper we present a neural network-based sensorless adaptive speed filter for induction motors. Only nameplate information and the actual motor currents and voltages are required for the initial setup of the proposed neural network-based speed filter. The speed filter gives acceptable steady state and transient speed response. The paper demonstrates the feasibility of adaptive speed filtering for induction motor which could be used for both diagnosis and control purposes.
{"title":"A neural network-based speed filter for induction motors: Adapting to motor load changes","authors":"R. Bharadwaj, A. Parlos, H. Toliyat, S. Menon","doi":"10.1109/IJCNN.1999.836219","DOIUrl":"https://doi.org/10.1109/IJCNN.1999.836219","url":null,"abstract":"Effective sensorless speed estimation is desirable for both online condition monitoring of induction motor and sensorless adjustable speed AC drive applications. In this paper we present a neural network-based sensorless adaptive speed filter for induction motors. Only nameplate information and the actual motor currents and voltages are required for the initial setup of the proposed neural network-based speed filter. The speed filter gives acceptable steady state and transient speed response. The paper demonstrates the feasibility of adaptive speed filtering for induction motor which could be used for both diagnosis and control purposes.","PeriodicalId":157719,"journal":{"name":"IJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No.99CH36339)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129624467","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 : 1999-07-10DOI: 10.1109/IJCNN.1999.833473
Eriko Ikeda, T. Imaoka, H. Ichihashi, T. Miyoshi
This paper proposes a new strategy of nonlinear component analysis for dimensionality reduction and representation of multidimensional data sets. The proposed procedure consists of two steps: one is to partition the data set into several clusters based on the local distances between two points, and the other is to project the obtained sub-manifolds on a low dimensional linear space by the multidimensional scaling methods.
{"title":"Nonlinear component analysis by fuzzy clustering and multidimensional scaling methods","authors":"Eriko Ikeda, T. Imaoka, H. Ichihashi, T. Miyoshi","doi":"10.1109/IJCNN.1999.833473","DOIUrl":"https://doi.org/10.1109/IJCNN.1999.833473","url":null,"abstract":"This paper proposes a new strategy of nonlinear component analysis for dimensionality reduction and representation of multidimensional data sets. The proposed procedure consists of two steps: one is to partition the data set into several clusters based on the local distances between two points, and the other is to project the obtained sub-manifolds on a low dimensional linear space by the multidimensional scaling methods.","PeriodicalId":157719,"journal":{"name":"IJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No.99CH36339)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129720032","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 : 1999-07-10DOI: 10.1109/IJCNN.1999.830871
A. Cangelosi
This paper uses a model of neural network and genetic algorithms to simulate the evolution of communication in populations of evolving neural networks. It focuses on the emergence of simple forms of syntax, i.e., the combination of two symbols. The simulation task resembles Savage-Rumbaugh and Rumbaugh's experiment (1978) on ape language and symbol acquisition. The simulation results show the evolution and cultural transmission of languages based on combination of grounded symbols. The model is analyzed according to the issues of the symbol grounding and symbol acquisition problems.
{"title":"Evolution of communication using symbol combination in populations of neural networks","authors":"A. Cangelosi","doi":"10.1109/IJCNN.1999.830871","DOIUrl":"https://doi.org/10.1109/IJCNN.1999.830871","url":null,"abstract":"This paper uses a model of neural network and genetic algorithms to simulate the evolution of communication in populations of evolving neural networks. It focuses on the emergence of simple forms of syntax, i.e., the combination of two symbols. The simulation task resembles Savage-Rumbaugh and Rumbaugh's experiment (1978) on ape language and symbol acquisition. The simulation results show the evolution and cultural transmission of languages based on combination of grounded symbols. The model is analyzed according to the issues of the symbol grounding and symbol acquisition problems.","PeriodicalId":157719,"journal":{"name":"IJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No.99CH36339)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129792710","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 : 1999-07-10DOI: 10.1109/IJCNN.1999.832665
R. Schultz, M. Hagan
We describe an online method of training neural networks, which is based on solving the linearized least-squares problem using the pseudo-inverse for the underdetermined case. This underdetermined linearized least squares (ULLS) method requires significantly less computation and memory for implementation than standard higher-order methods such as the Gauss-Newton method or extended Kalman filter. This decrease is possible because the method allows training to proceed with a smaller number of samples than parameters. Simulation results which compare the performance of the ULLS algorithm to the recursive linearized least squares algorithm (RLLS) and the gradient descent algorithm are presented. Results showing the impact on computational complexity and squared-error performance of the ULLS method, when the number of terms in the Jacobian matrix is varied, are also presented.
{"title":"Online least-squares training for the underdetermined case","authors":"R. Schultz, M. Hagan","doi":"10.1109/IJCNN.1999.832665","DOIUrl":"https://doi.org/10.1109/IJCNN.1999.832665","url":null,"abstract":"We describe an online method of training neural networks, which is based on solving the linearized least-squares problem using the pseudo-inverse for the underdetermined case. This underdetermined linearized least squares (ULLS) method requires significantly less computation and memory for implementation than standard higher-order methods such as the Gauss-Newton method or extended Kalman filter. This decrease is possible because the method allows training to proceed with a smaller number of samples than parameters. Simulation results which compare the performance of the ULLS algorithm to the recursive linearized least squares algorithm (RLLS) and the gradient descent algorithm are presented. Results showing the impact on computational complexity and squared-error performance of the ULLS method, when the number of terms in the Jacobian matrix is varied, are also presented.","PeriodicalId":157719,"journal":{"name":"IJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No.99CH36339)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129934046","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 : 1999-07-10DOI: 10.1109/IJCNN.1999.830852
M. J. McAlister, S. Wermter
HyValue is a hybrid electronic submission system which utilizes techniques from natural language processing, neural networks and rule based systems to accept, evaluate and mark work submitted by a student for reading or writing. This paper describes the theory behind the system design and the development of the individual components and their interaction. Issues addressed include the definition of sentence structure, fuzzy rule construction and integration with a knowledge base containing the marking rubrics for reading and writing. An evaluation of the system is provided and conclusions drawn.
{"title":"Rule generation from neural networks for student assessment","authors":"M. J. McAlister, S. Wermter","doi":"10.1109/IJCNN.1999.830852","DOIUrl":"https://doi.org/10.1109/IJCNN.1999.830852","url":null,"abstract":"HyValue is a hybrid electronic submission system which utilizes techniques from natural language processing, neural networks and rule based systems to accept, evaluate and mark work submitted by a student for reading or writing. This paper describes the theory behind the system design and the development of the individual components and their interaction. Issues addressed include the definition of sentence structure, fuzzy rule construction and integration with a knowledge base containing the marking rubrics for reading and writing. An evaluation of the system is provided and conclusions drawn.","PeriodicalId":157719,"journal":{"name":"IJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No.99CH36339)","volume":"06 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129944376","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 : 1999-07-10DOI: 10.1109/IJCNN.1999.833537
Yong Liu, Bin Zhao, Shaowei Xia, Ming-Sheng Zhao
This paper proposes a self-organizing network with the fuzzy hyperellipsoid-classifier (FHECFN) and utilizes it to recognize handwritten numerals. Based on the clustering result of SOM, FHECFN divides the center that performs worse taking the advantage of the fuzzy hyperellipsoidal clustering algorithm. When reaching the satisfying requirement, the network stops divining and then obtains the suitable number of prototypes and the hyperellipsoidal classifying result. With the supervised learning algorithm, such as learning vector quantization, the network achieves a better learning result and in the experiments of recognizing the handwritten numerals, the network shows a promising performance.
{"title":"A self-organizing network with fuzzy hyperellipsoidal classifying and its application in handwritten numeral recognition","authors":"Yong Liu, Bin Zhao, Shaowei Xia, Ming-Sheng Zhao","doi":"10.1109/IJCNN.1999.833537","DOIUrl":"https://doi.org/10.1109/IJCNN.1999.833537","url":null,"abstract":"This paper proposes a self-organizing network with the fuzzy hyperellipsoid-classifier (FHECFN) and utilizes it to recognize handwritten numerals. Based on the clustering result of SOM, FHECFN divides the center that performs worse taking the advantage of the fuzzy hyperellipsoidal clustering algorithm. When reaching the satisfying requirement, the network stops divining and then obtains the suitable number of prototypes and the hyperellipsoidal classifying result. With the supervised learning algorithm, such as learning vector quantization, the network achieves a better learning result and in the experiments of recognizing the handwritten numerals, the network shows a promising performance.","PeriodicalId":157719,"journal":{"name":"IJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No.99CH36339)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128470428","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 : 1999-07-10DOI: 10.1109/IJCNN.1999.830769
S. Chi, Hung-Pin Chen, Chun-Hao Cheng
Previously used quantitative indices for predicting stock prices are not really suitable, and the requirement for a large amount of input data slows down the convergence of a neural network model. Therefore, this research attempts to develop a better prediction model by the integration of neural network technique and grey theory for the SIMEX Taiwan stock index future. In this research, the grey theory applied include grey forecast model and grey relationship analysis. The grey forecast model, GM(1,1), was applied to predict the next day's stock index future. To examine the influence of dimension of the model to prediction accuracy, seven different kinds of dimension 5, 6, 8, 10, 12, 14, and 15 were tested. The generated data were then regarded as new technical indices in grey relationship analysis and prediction of neural network. Grey relationship analysis was used to filter the most important quantitative technical indices. Finally, a recurrent neural network was developed to train and predict the price trend of stock index future. In the network structure, the price trend of stock index future is the output and the values gained from previous processing in grey relationship analysis is the input. The conclusion shows our models can provide good prediction for this problem.
{"title":"A forecasting approach for stock index future using grey theory and neural networks","authors":"S. Chi, Hung-Pin Chen, Chun-Hao Cheng","doi":"10.1109/IJCNN.1999.830769","DOIUrl":"https://doi.org/10.1109/IJCNN.1999.830769","url":null,"abstract":"Previously used quantitative indices for predicting stock prices are not really suitable, and the requirement for a large amount of input data slows down the convergence of a neural network model. Therefore, this research attempts to develop a better prediction model by the integration of neural network technique and grey theory for the SIMEX Taiwan stock index future. In this research, the grey theory applied include grey forecast model and grey relationship analysis. The grey forecast model, GM(1,1), was applied to predict the next day's stock index future. To examine the influence of dimension of the model to prediction accuracy, seven different kinds of dimension 5, 6, 8, 10, 12, 14, and 15 were tested. The generated data were then regarded as new technical indices in grey relationship analysis and prediction of neural network. Grey relationship analysis was used to filter the most important quantitative technical indices. Finally, a recurrent neural network was developed to train and predict the price trend of stock index future. In the network structure, the price trend of stock index future is the output and the values gained from previous processing in grey relationship analysis is the input. The conclusion shows our models can provide good prediction for this problem.","PeriodicalId":157719,"journal":{"name":"IJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No.99CH36339)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128559077","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 : 1999-07-10DOI: 10.1109/IJCNN.1999.833411
Chih-Chi Chang, Chungyong Tsai
For engineering applications, complex pre-calculations may be necessary to evaluate the fuzzy degree of a simple linguistic term in the conditional part. If these pre-calculations are excluded from the system, it is difficult to discriminate the physical meaning of the rule from the neural nets. Therefore, this work applies feature extraction to replace these pre-calculations. The proposed neural fuzzy system extracts features by heterogeneous neurons. The proposed system's structure and its advantages are described in detailed. A controller designed by the proposed neural fuzzy system is presented as well.
{"title":"A neural-fuzzy controller with heterogeneous neurons","authors":"Chih-Chi Chang, Chungyong Tsai","doi":"10.1109/IJCNN.1999.833411","DOIUrl":"https://doi.org/10.1109/IJCNN.1999.833411","url":null,"abstract":"For engineering applications, complex pre-calculations may be necessary to evaluate the fuzzy degree of a simple linguistic term in the conditional part. If these pre-calculations are excluded from the system, it is difficult to discriminate the physical meaning of the rule from the neural nets. Therefore, this work applies feature extraction to replace these pre-calculations. The proposed neural fuzzy system extracts features by heterogeneous neurons. The proposed system's structure and its advantages are described in detailed. A controller designed by the proposed neural fuzzy system is presented as well.","PeriodicalId":157719,"journal":{"name":"IJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No.99CH36339)","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128685751","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 : 1999-07-10DOI: 10.1109/IJCNN.1999.836057
R. Kuo, S. Chi, B. W. Den
This paper proposes a novel fuzzy neural network for clustering the parts into several families. The proposed network, which has fuzzy inputs as well as fuzzy weights, integrates the Kohonen's feature map neural network and the fuzzy set theory. The model evaluation results show that the proposed fuzzy neural network can provide more accurate decision compared to the fuzzy c-means algorithm and k-means algorithm.
{"title":"A fuzzy Kohonen's feature map neural network with application to group technology","authors":"R. Kuo, S. Chi, B. W. Den","doi":"10.1109/IJCNN.1999.836057","DOIUrl":"https://doi.org/10.1109/IJCNN.1999.836057","url":null,"abstract":"This paper proposes a novel fuzzy neural network for clustering the parts into several families. The proposed network, which has fuzzy inputs as well as fuzzy weights, integrates the Kohonen's feature map neural network and the fuzzy set theory. The model evaluation results show that the proposed fuzzy neural network can provide more accurate decision compared to the fuzzy c-means algorithm and k-means algorithm.","PeriodicalId":157719,"journal":{"name":"IJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No.99CH36339)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128704151","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：