Pub Date : 1989-11-14DOI: 10.1109/ICSMC.1989.71324
A. Calero
A control process is proposed based on managing a large number of influence parameters; using clustering/visualization concepts to manage the complexity associated with such large amounts of information; and considering as a reference the global knowledge of a distributed human control network. It is emphasized that using the strong capabilities of human visual reasoning in the analysis of complex systems may be an effective way of dealing with their complexity. An effective visualization process based on geometric 2-D/3-D representations of multidimensional spaces as well as clustering procedures constitutes the key to the process. Defining an appropriate reference model is also important and possible if a deep knowledge of the system is available. A simple demonstration model developed to evaluate some ideas of the proposed system is described.<>
{"title":"Managing complexity in large corporate information systems","authors":"A. Calero","doi":"10.1109/ICSMC.1989.71324","DOIUrl":"https://doi.org/10.1109/ICSMC.1989.71324","url":null,"abstract":"A control process is proposed based on managing a large number of influence parameters; using clustering/visualization concepts to manage the complexity associated with such large amounts of information; and considering as a reference the global knowledge of a distributed human control network. It is emphasized that using the strong capabilities of human visual reasoning in the analysis of complex systems may be an effective way of dealing with their complexity. An effective visualization process based on geometric 2-D/3-D representations of multidimensional spaces as well as clustering procedures constitutes the key to the process. Defining an appropriate reference model is also important and possible if a deep knowledge of the system is available. A simple demonstration model developed to evaluate some ideas of the proposed system is described.<<ETX>>","PeriodicalId":72691,"journal":{"name":"Conference proceedings. IEEE International Conference on Systems, Man, and Cybernetics","volume":"1 1","pages":"393-398 vol.1"},"PeriodicalIF":0.0,"publicationDate":"1989-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86524560","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 : 1989-11-14DOI: 10.1109/ICSMC.1989.71280
L. Perlovsky
A maximal-likelihood artificial neural system (MLANS) is described which performs the ML classification for problems requiring nonlinear classification boundaries. This neural network has ML neurons, which adaptively estimate the local metric in the classification space. This permits the design of flexible classifier shapes using a no-hidden-layer architecture and provides orders-of-magnitude improvement in learning efficiency. The learning efficiency of this network approaches the Cramer-Rao bounds with a relatively small number of samples. The learning process of MLANS can be unsupervised learning with partial or imperfect supervision. The ML approach allows for optimal fusion of all available information, such as a priori and real-time information, including supervisory (training) information.<>
{"title":"Neural networks with maximal adaptive efficiency","authors":"L. Perlovsky","doi":"10.1109/ICSMC.1989.71280","DOIUrl":"https://doi.org/10.1109/ICSMC.1989.71280","url":null,"abstract":"A maximal-likelihood artificial neural system (MLANS) is described which performs the ML classification for problems requiring nonlinear classification boundaries. This neural network has ML neurons, which adaptively estimate the local metric in the classification space. This permits the design of flexible classifier shapes using a no-hidden-layer architecture and provides orders-of-magnitude improvement in learning efficiency. The learning efficiency of this network approaches the Cramer-Rao bounds with a relatively small number of samples. The learning process of MLANS can be unsupervised learning with partial or imperfect supervision. The ML approach allows for optimal fusion of all available information, such as a priori and real-time information, including supervisory (training) information.<<ETX>>","PeriodicalId":72691,"journal":{"name":"Conference proceedings. IEEE International Conference on Systems, Man, and Cybernetics","volume":"70 1","pages":"208-209 vol.1"},"PeriodicalIF":0.0,"publicationDate":"1989-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83909414","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 : 1989-11-14DOI: 10.1109/ICSMC.1989.71349
R.Y. Lorin, Pierre Morizet-Mahoudeaux
The problem of using global variables to improve the representation capacities of the SUPER expert system is addressed. The first difficulty is to propagate the domain of definition of the variables along the knowledge base using the AND/OR transitive relations between rules. The second problem is to upgrade the algorithm that has been defined for the consistency maintenance of the knowledge base for propositional logic to the level of logic with global variables. Answers are given to both problems: (1) the principles for chaining rules with global variables are given; (2) tools for computing the propagation of the domain of variables along the knowledge base are defined; (3) consistency definitions are given and tools for maintaining the consistency of the knowledge base as it incrementally increases are proposed.<>
{"title":"Handling variables in a logical network of rules, revision of consistency controlling method of the plan of action manager SUPER","authors":"R.Y. Lorin, Pierre Morizet-Mahoudeaux","doi":"10.1109/ICSMC.1989.71349","DOIUrl":"https://doi.org/10.1109/ICSMC.1989.71349","url":null,"abstract":"The problem of using global variables to improve the representation capacities of the SUPER expert system is addressed. The first difficulty is to propagate the domain of definition of the variables along the knowledge base using the AND/OR transitive relations between rules. The second problem is to upgrade the algorithm that has been defined for the consistency maintenance of the knowledge base for propositional logic to the level of logic with global variables. Answers are given to both problems: (1) the principles for chaining rules with global variables are given; (2) tools for computing the propagation of the domain of variables along the knowledge base are defined; (3) consistency definitions are given and tools for maintaining the consistency of the knowledge base as it incrementally increases are proposed.<<ETX>>","PeriodicalId":72691,"journal":{"name":"Conference proceedings. IEEE International Conference on Systems, Man, and Cybernetics","volume":"67 1","pages":"516-520 vol.2"},"PeriodicalIF":0.0,"publicationDate":"1989-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89024502","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 : 1989-11-14DOI: 10.1109/ICSMC.1989.71390
Abhinandan Jain, P. Gonsalves, G. Zacharias
A general formulation of the optimal control model of the human operator is presented. This formulation dispenses with the assumption that the operator has an accurate internal model of the system dynamics and other parameters affecting his control behavior. It has applications to learning and transfer of training situations where the operator refines his behavior over a period of time. Validation of the model is via analysis of a flight control experiment using a discrete transfer of the training protocol.<>
{"title":"Internal model mismatch in the optimal control model","authors":"Abhinandan Jain, P. Gonsalves, G. Zacharias","doi":"10.1109/ICSMC.1989.71390","DOIUrl":"https://doi.org/10.1109/ICSMC.1989.71390","url":null,"abstract":"A general formulation of the optimal control model of the human operator is presented. This formulation dispenses with the assumption that the operator has an accurate internal model of the system dynamics and other parameters affecting his control behavior. It has applications to learning and transfer of training situations where the operator refines his behavior over a period of time. Validation of the model is via analysis of a flight control experiment using a discrete transfer of the training protocol.<<ETX>>","PeriodicalId":72691,"journal":{"name":"Conference proceedings. IEEE International Conference on Systems, Man, and Cybernetics","volume":"11 1","pages":"731-737 vol.2"},"PeriodicalIF":0.0,"publicationDate":"1989-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88710901","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 : 1989-11-14DOI: 10.1109/ICSMC.1989.71323
E. Khalimsky
Large-scale management information systems, knowledge-based systems, neural systems, control systems and others are becoming extremely sophisticated. In order to investigate their hierarchical structure the author considers such systems as topological spaces. These systems are partitioned into numerous subsystems, cells or elements in the process of their analysis and design. Such decomposition is equivalent to representation of the corresponding topological space as a disjoint union of cells. A topological cell complex of some space X is built as a quotient space of X, which is constructed by representation of each cell as one point. Thus cell complexes and CW-complexes of topological spaces are considered as topological spaces with quotient topology. Such topological complexes contain fewer points than corresponding spaces, and preserve their structure. This feature makes them useful for structural systems analysis and design.<>
{"title":"Topological complexes in analysis and design of large scale system-factorization methodology","authors":"E. Khalimsky","doi":"10.1109/ICSMC.1989.71323","DOIUrl":"https://doi.org/10.1109/ICSMC.1989.71323","url":null,"abstract":"Large-scale management information systems, knowledge-based systems, neural systems, control systems and others are becoming extremely sophisticated. In order to investigate their hierarchical structure the author considers such systems as topological spaces. These systems are partitioned into numerous subsystems, cells or elements in the process of their analysis and design. Such decomposition is equivalent to representation of the corresponding topological space as a disjoint union of cells. A topological cell complex of some space X is built as a quotient space of X, which is constructed by representation of each cell as one point. Thus cell complexes and CW-complexes of topological spaces are considered as topological spaces with quotient topology. Such topological complexes contain fewer points than corresponding spaces, and preserve their structure. This feature makes them useful for structural systems analysis and design.<<ETX>>","PeriodicalId":72691,"journal":{"name":"Conference proceedings. IEEE International Conference on Systems, Man, and Cybernetics","volume":"13 1","pages":"390-392 vol.1"},"PeriodicalIF":0.0,"publicationDate":"1989-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90596311","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 : 1989-11-14DOI: 10.1109/ICSMC.1989.71298
I. D'Antone, L. Fortuna, G. Nunnari
The implementation of an expert system making possible online fault diagnosis on a complex data acquisition system, a particle detector, designed for the MACRO (Monopoles, Astrophysics and Cosmic Ray Observatory) physics investigations is outlined. The system, called MADIES (Macro Diagnostic Expert System), is designed to take appropriate control actions in order to reduce the unavailability of the apparatus. MADIES was developed using the NEXPERT commercial shell.<>
{"title":"An intelligent diagnosis and control system for a complex data acquisition equipment in nuclear physics investigations","authors":"I. D'Antone, L. Fortuna, G. Nunnari","doi":"10.1109/ICSMC.1989.71298","DOIUrl":"https://doi.org/10.1109/ICSMC.1989.71298","url":null,"abstract":"The implementation of an expert system making possible online fault diagnosis on a complex data acquisition system, a particle detector, designed for the MACRO (Monopoles, Astrophysics and Cosmic Ray Observatory) physics investigations is outlined. The system, called MADIES (Macro Diagnostic Expert System), is designed to take appropriate control actions in order to reduce the unavailability of the apparatus. MADIES was developed using the NEXPERT commercial shell.<<ETX>>","PeriodicalId":72691,"journal":{"name":"Conference proceedings. IEEE International Conference on Systems, Man, and Cybernetics","volume":"1 1","pages":"282-285 vol.1"},"PeriodicalIF":0.0,"publicationDate":"1989-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90715798","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 : 1989-11-14DOI: 10.1109/ICSMC.1989.71279
R. Allen
A reinforcement training procedure was developed for sequential back-propagation networks and applied in several studies demonstrating interaction between agents in multiple-agent networks. In the first study, a network was trained to predict the next position of an agent which was moving in a complex pattern around the corners of a square. The network quickly learned to predict the position without error. In particular, the network may be said to have developed an agent or user model of the moving agent. In two additional studies, a joint contingency was applied to two agents and limited cooperation was developed between them. Overall, the results provide support for the application of neural networks in distributed AI (artificial intelligence).<>
{"title":"Developing agent models with a neural reinforcement technique","authors":"R. Allen","doi":"10.1109/ICSMC.1989.71279","DOIUrl":"https://doi.org/10.1109/ICSMC.1989.71279","url":null,"abstract":"A reinforcement training procedure was developed for sequential back-propagation networks and applied in several studies demonstrating interaction between agents in multiple-agent networks. In the first study, a network was trained to predict the next position of an agent which was moving in a complex pattern around the corners of a square. The network quickly learned to predict the position without error. In particular, the network may be said to have developed an agent or user model of the moving agent. In two additional studies, a joint contingency was applied to two agents and limited cooperation was developed between them. Overall, the results provide support for the application of neural networks in distributed AI (artificial intelligence).<<ETX>>","PeriodicalId":72691,"journal":{"name":"Conference proceedings. IEEE International Conference on Systems, Man, and Cybernetics","volume":"49 1","pages":"206-207 vol.1"},"PeriodicalIF":0.0,"publicationDate":"1989-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90991116","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 : 1989-11-14DOI: 10.1109/ICSMC.1989.71439
C. Lewis
Operators develop habitual responses for controlling complex systems which may be neither apparent nor verbalized. Identifying consistent patterns in their behavior can reveal these implicit strategies. Logs from a simulated process plant were used to identify production rules characterizing operators' behavior. Similarity among strategies was determined by finding the intersections and unions among the preconditions of these rules. Clusters of subjects using similar strategies were identified and compared on performance measures. This characterization of strategies provided a good account of differences in performance which were readily interpretable in terms of the task. The approach appears promising as a technique for diagnosing performance in simulator training.<>
{"title":"Comparing subject strategies at a process control task","authors":"C. Lewis","doi":"10.1109/ICSMC.1989.71439","DOIUrl":"https://doi.org/10.1109/ICSMC.1989.71439","url":null,"abstract":"Operators develop habitual responses for controlling complex systems which may be neither apparent nor verbalized. Identifying consistent patterns in their behavior can reveal these implicit strategies. Logs from a simulated process plant were used to identify production rules characterizing operators' behavior. Similarity among strategies was determined by finding the intersections and unions among the preconditions of these rules. Clusters of subjects using similar strategies were identified and compared on performance measures. This characterization of strategies provided a good account of differences in performance which were readily interpretable in terms of the task. The approach appears promising as a technique for diagnosing performance in simulator training.<<ETX>>","PeriodicalId":72691,"journal":{"name":"Conference proceedings. IEEE International Conference on Systems, Man, and Cybernetics","volume":"32 1","pages":"971-976 vol.3"},"PeriodicalIF":0.0,"publicationDate":"1989-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89545394","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 : 1989-11-14DOI: 10.1109/ICSMC.1989.71451
P. Lim, J. Tien
In a distributed environment where data are distributed and stored in individual DASDs, the performance of each of these DASD subsystems is critical to overall system performance. Hence, in order to size the DASD subsystems within a distributed environment appropriately, a flexible and robust planning tool is needed. A planning tool in the form of a robust simulation-based model is proposed for the performance evaluation of noncached and cached DASD subsystems. It is validated with performance data obtained from a real-world environment. This planning tool is part of an ongoing research project in the development of a four-level distributed information system simulator.<>
{"title":"Direct access storage device (DASD) modeling and validation","authors":"P. Lim, J. Tien","doi":"10.1109/ICSMC.1989.71451","DOIUrl":"https://doi.org/10.1109/ICSMC.1989.71451","url":null,"abstract":"In a distributed environment where data are distributed and stored in individual DASDs, the performance of each of these DASD subsystems is critical to overall system performance. Hence, in order to size the DASD subsystems within a distributed environment appropriately, a flexible and robust planning tool is needed. A planning tool in the form of a robust simulation-based model is proposed for the performance evaluation of noncached and cached DASD subsystems. It is validated with performance data obtained from a real-world environment. This planning tool is part of an ongoing research project in the development of a four-level distributed information system simulator.<<ETX>>","PeriodicalId":72691,"journal":{"name":"Conference proceedings. IEEE International Conference on Systems, Man, and Cybernetics","volume":"8 1","pages":"1024-1029 vol.3"},"PeriodicalIF":0.0,"publicationDate":"1989-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89551702","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 : 1989-11-14DOI: 10.1109/ICSMC.1989.71452
J. Gammack, S. Battle, R. Stephens
A knowledge acquisition and representation scheme is introduced which is suited both to serial processing in production systems and to a constraint-based representation evaluated in parallel. Two applications are described: a rule-based system for financial risk assessment in life insurance underwriting, and a decision-support aid for energy management in a steel plant. These have been programmed both in conventional serial languages and in Occam on a transputer.<>
{"title":"A knowledge acquisition and representation scheme for constraint based and parallel systems","authors":"J. Gammack, S. Battle, R. Stephens","doi":"10.1109/ICSMC.1989.71452","DOIUrl":"https://doi.org/10.1109/ICSMC.1989.71452","url":null,"abstract":"A knowledge acquisition and representation scheme is introduced which is suited both to serial processing in production systems and to a constraint-based representation evaluated in parallel. Two applications are described: a rule-based system for financial risk assessment in life insurance underwriting, and a decision-support aid for energy management in a steel plant. These have been programmed both in conventional serial languages and in Occam on a transputer.<<ETX>>","PeriodicalId":72691,"journal":{"name":"Conference proceedings. IEEE International Conference on Systems, Man, and Cybernetics","volume":"599 1","pages":"1030-1035 vol.3"},"PeriodicalIF":0.0,"publicationDate":"1989-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76791006","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：