Pub Date : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018111
P. Amato, A. di Nola, B. Gerla
We describe a correspondence between rational Lukasiewicz formulas and neural networks in which the activation function is the truncated identity and synaptic weights are rational numbers. On one hand, having a logical representation (in a given logic) of neural networks could widen the interpretability, amalgamability and reuse of these objects. On the other hand, neural networks could be used to learn formulas from data and as circuital counterparts of (functions represented by) formulas.
{"title":"Neural networks and rational Lukasiewicz logic","authors":"P. Amato, A. di Nola, B. Gerla","doi":"10.1109/NAFIPS.2002.1018111","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018111","url":null,"abstract":"We describe a correspondence between rational Lukasiewicz formulas and neural networks in which the activation function is the truncated identity and synaptic weights are rational numbers. On one hand, having a logical representation (in a given logic) of neural networks could widen the interpretability, amalgamability and reuse of these objects. On the other hand, neural networks could be used to learn formulas from data and as circuital counterparts of (functions represented by) formulas.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"129 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116537339","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 : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018074
T. Young, T.Y. Lin
A rough membership function uses counting probability (ratio of cardinal numbers) to define a membership. An extension, called granular membership function (GMF), generalizes the counting probability to a general set function (GSF), such as probability, possibility, belief function, etc. have been investigated previously. The "set theoretical operations" (STO) of GMF are induced naturally from the operations of GSF. In particular, probabilistic GMF (PGMF) are defined according to the rules of probability; their operations depend not only on the numerical grades but also on the events. This is often expressed as "STO are not truth functional." On the other hand, STO on traditional fuzzy sets are truth functional. This phenomenon led us to conclude the grade of traditional fuzzy sets can not be interpreted as a probability.
{"title":"Fuzzy sets, rough set and probability","authors":"T. Young, T.Y. Lin","doi":"10.1109/NAFIPS.2002.1018074","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018074","url":null,"abstract":"A rough membership function uses counting probability (ratio of cardinal numbers) to define a membership. An extension, called granular membership function (GMF), generalizes the counting probability to a general set function (GSF), such as probability, possibility, belief function, etc. have been investigated previously. The \"set theoretical operations\" (STO) of GMF are induced naturally from the operations of GSF. In particular, probabilistic GMF (PGMF) are defined according to the rules of probability; their operations depend not only on the numerical grades but also on the events. This is often expressed as \"STO are not truth functional.\" On the other hand, STO on traditional fuzzy sets are truth functional. This phenomenon led us to conclude the grade of traditional fuzzy sets can not be interpreted as a probability.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"80 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132191946","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 : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018125
D. van Cleave, K. Rattan
Control law partitioning is a widely used concept that incorporates a mathematical model of the plant into the control system. This is both an advantage and disadvantage. With an accurate model, the system control is much more robust and easy to manage. However, with a complex nonlinear system, an accurate mathematical model can be very difficult to obtain. A fuzzy logic controller can be developed that makes use of empirically derived data thereby accurately modeling the plant without the necessity of a mathematical model. Tuning such a controller to the empirical data can be problematic, so a tuning algorithm is used to adjust the controller parameters for optimal performance. In this paper, a fourth-order system is used as a demonstration plant, a fuzzy logic controller is developed using a very fast offline tuning algorithm, and the performance of the resulting controller is examined.
{"title":"Control law partitioning via fuzzy logic control","authors":"D. van Cleave, K. Rattan","doi":"10.1109/NAFIPS.2002.1018125","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018125","url":null,"abstract":"Control law partitioning is a widely used concept that incorporates a mathematical model of the plant into the control system. This is both an advantage and disadvantage. With an accurate model, the system control is much more robust and easy to manage. However, with a complex nonlinear system, an accurate mathematical model can be very difficult to obtain. A fuzzy logic controller can be developed that makes use of empirically derived data thereby accurately modeling the plant without the necessity of a mathematical model. Tuning such a controller to the empirical data can be problematic, so a tuning algorithm is used to adjust the controller parameters for optimal performance. In this paper, a fourth-order system is used as a demonstration plant, a fuzzy logic controller is developed using a very fast offline tuning algorithm, and the performance of the resulting controller is examined.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132053689","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 : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018097
L. Legey, H. T. Firmo
Proposes a confluence between soft OR and soft computing methods, by means of an application of fuzzy logic ideas to robustness analysis. Both methods try to add flexibility to proposed solutions, of real world problems, although insofar as their application contexts is concerned, they have been quite apart. So it is only natural to bring them together in what could be called a "fuzzification" of robustness analysis. A first step in that direction is attempted in the paper.
{"title":"Fuzzy systems and soft O.R","authors":"L. Legey, H. T. Firmo","doi":"10.1109/NAFIPS.2002.1018097","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018097","url":null,"abstract":"Proposes a confluence between soft OR and soft computing methods, by means of an application of fuzzy logic ideas to robustness analysis. Both methods try to add flexibility to proposed solutions, of real world problems, although insofar as their application contexts is concerned, they have been quite apart. So it is only natural to bring them together in what could be called a \"fuzzification\" of robustness analysis. A first step in that direction is attempted in the paper.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114683524","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 : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018046
T. Martin
The vision of a semantic Web incorporates many aspects which require flexible knowledge representation, learning and reasoning. These include: the mismatch between crisp hierarchical structures and the 'fuzzier" real world in which objects may have partial membership in classes; notions of approximate equality in data, and semantic equivalence of syntactically different structures; and robustness against inconsistent, missing, partial or incorrect data. In this paper we outline a system which uses concept hierarchies to focus queries. We concentrate on the need to avoid rigid definitions and allow uncertainty in the concept hierarchy, in order to combine diverse data sources.
{"title":"Softer concepts mean smarter queries","authors":"T. Martin","doi":"10.1109/NAFIPS.2002.1018046","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018046","url":null,"abstract":"The vision of a semantic Web incorporates many aspects which require flexible knowledge representation, learning and reasoning. These include: the mismatch between crisp hierarchical structures and the 'fuzzier\" real world in which objects may have partial membership in classes; notions of approximate equality in data, and semantic equivalence of syntactically different structures; and robustness against inconsistent, missing, partial or incorrect data. In this paper we outline a system which uses concept hierarchies to focus queries. We concentrate on the need to avoid rigid definitions and allow uncertainty in the concept hierarchy, in order to combine diverse data sources.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"173 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132762301","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 : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018041
J. Tomé, Joao Paulo Carvalho
Fuzzy Boolean Networks are Boolean networks with nature like characteristics, such as organization of neurons on cards or areas. random individual connections, structured meshes of links between cards. They also share with natural systems some interesting properties: relative noise immunity, capability of approximate reasoning and learning from sets of experiments. An interesting problem related with these nets is the number of different rules that they are able to capture front experiments, that is, their rule capacity. This work establishes a lower bound for this number, proving that it depends on the number of inputs per consequent neurons.
{"title":"Rule capacity in fuzzy boolean networks","authors":"J. Tomé, Joao Paulo Carvalho","doi":"10.1109/NAFIPS.2002.1018041","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018041","url":null,"abstract":"Fuzzy Boolean Networks are Boolean networks with nature like characteristics, such as organization of neurons on cards or areas. random individual connections, structured meshes of links between cards. They also share with natural systems some interesting properties: relative noise immunity, capability of approximate reasoning and learning from sets of experiments. An interesting problem related with these nets is the number of different rules that they are able to capture front experiments, that is, their rule capacity. This work establishes a lower bound for this number, proving that it depends on the number of inputs per consequent neurons.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116796188","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 : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018062
Hamid R. Tizhoosh
In the literature, there already exist some fuzzy approaches to edge detection. However, they are generally computationally expensive. In this paper, several fast fuzzy edge detectors are introduced for practical cases where a rough edge map is needed in a short time.
{"title":"Fast fuzzy edge detection","authors":"Hamid R. Tizhoosh","doi":"10.1109/NAFIPS.2002.1018062","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018062","url":null,"abstract":"In the literature, there already exist some fuzzy approaches to edge detection. However, they are generally computationally expensive. In this paper, several fast fuzzy edge detectors are introduced for practical cases where a rough edge map is needed in a short time.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125198923","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 : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018040
C. Helgason, T. Jobe
Background: Current clinical trials in medicine use probability-based statistics. Statistics separates the patient's physiologic elements from his body and claims causal correlation. Methods: In this study we derive measures of necessary and sufficient causal ground, formal causal ground and clinical causal effect from the fuzzy subsethood theorem as defined by Kosko. We represent patients as sets as points in a unit hypercube before and after treatment with antiplatelet agents. Results: The measures of formal causal ground and clinical causal effect are in units of cardinality. Using data from 16 patients taking antiplatelet therapy, we derived formal causal ground and clinical causal effect which in an imaginary clinical FCM represent causal edge strengths for nodes of antiplatelet medication. Conclusion: Our causal measures are represented as changes in cardinality in a unit hypercube and can be used instead of probability based statistics to judge the causal relation of medical therapies or conditions.
{"title":"Necessary and sufficient causal ground and effect is measured by fuzzy cardinality and may represent natural edge strength connections in a clinical fuzzy cognitive map","authors":"C. Helgason, T. Jobe","doi":"10.1109/NAFIPS.2002.1018040","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018040","url":null,"abstract":"Background: Current clinical trials in medicine use probability-based statistics. Statistics separates the patient's physiologic elements from his body and claims causal correlation. Methods: In this study we derive measures of necessary and sufficient causal ground, formal causal ground and clinical causal effect from the fuzzy subsethood theorem as defined by Kosko. We represent patients as sets as points in a unit hypercube before and after treatment with antiplatelet agents. Results: The measures of formal causal ground and clinical causal effect are in units of cardinality. Using data from 16 patients taking antiplatelet therapy, we derived formal causal ground and clinical causal effect which in an imaginary clinical FCM represent causal edge strengths for nodes of antiplatelet medication. Conclusion: Our causal measures are represented as changes in cardinality in a unit hypercube and can be used instead of probability based statistics to judge the causal relation of medical therapies or conditions.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"27 106 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125932352","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 : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018115
L. A. Zadeh
Existing search engines have many remarkable capabilities. But what is not among them is deduction capability-the capability to answer a query by drawing on information which resides in various parts of the knowledge base or is augmented by the user. The problem-which is not widely recognized-is that much of the information in the knowledge base of a search engine is perception-based. Methods based on bivalent logic and standard probability theory lack capability to operate on perception-based information. A search engine with deduction capability is, in effect, a question-answering system. Limited progress toward a realization of deduction capability is achievable through application of methods based on bivalent logic and standard probability theory. But to move beyond the reach of standard methods it is necessary to change direction. In the approach which is outlined, a concept which plays a pivotal role is that of a prototype-a concept which has a position of centrality in human reasoning, recognition, search and decision processes.
{"title":"A prototype-centered approach to adding deduction capability to search engines-the concept of protoform","authors":"L. A. Zadeh","doi":"10.1109/NAFIPS.2002.1018115","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018115","url":null,"abstract":"Existing search engines have many remarkable capabilities. But what is not among them is deduction capability-the capability to answer a query by drawing on information which resides in various parts of the knowledge base or is augmented by the user. The problem-which is not widely recognized-is that much of the information in the knowledge base of a search engine is perception-based. Methods based on bivalent logic and standard probability theory lack capability to operate on perception-based information. A search engine with deduction capability is, in effect, a question-answering system. Limited progress toward a realization of deduction capability is achievable through application of methods based on bivalent logic and standard probability theory. But to move beyond the reach of standard methods it is necessary to change direction. In the approach which is outlined, a concept which plays a pivotal role is that of a prototype-a concept which has a position of centrality in human reasoning, recognition, search and decision processes.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128882023","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 : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018056
Hugang Han
Based on the Lyapunov synthesis approach and regarding the fuzzy systems as approximators to approximate the unknown functions in the system to be controlled, several adaptive fuzzy control schemes have been developed during the last decade. Actually, these schemes have been applied only to simple classes of nonlinear systems. In the concrete, (i) most of them just consider SISO systems (which can avoid the challenging of the coupling between control inputs); (ii) the upper bounds of uncertainties, and the reconstruction errors between the best approximators and their corresponding functions to be approximated are assumed to be known (in this way, the traditional adaptive methods or robust methods could be utilized straightforwardly). This paper develops a design methodology that expends the class of nonlinear systems to MIMO systems, the above restrictive assumptions can be relaxed by using an unique way to deal with the uncertainties and the reconstruction errors. The overall adaptive scheme is shown to guarantee the tracking error, between the outputs of system and the desired values, to be asymptotical in decay.
{"title":"Adaptive fuzzy control scheme for MIMO systems with uncertainties","authors":"Hugang Han","doi":"10.1109/NAFIPS.2002.1018056","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018056","url":null,"abstract":"Based on the Lyapunov synthesis approach and regarding the fuzzy systems as approximators to approximate the unknown functions in the system to be controlled, several adaptive fuzzy control schemes have been developed during the last decade. Actually, these schemes have been applied only to simple classes of nonlinear systems. In the concrete, (i) most of them just consider SISO systems (which can avoid the challenging of the coupling between control inputs); (ii) the upper bounds of uncertainties, and the reconstruction errors between the best approximators and their corresponding functions to be approximated are assumed to be known (in this way, the traditional adaptive methods or robust methods could be utilized straightforwardly). This paper develops a design methodology that expends the class of nonlinear systems to MIMO systems, the above restrictive assumptions can be relaxed by using an unique way to deal with the uncertainties and the reconstruction errors. The overall adaptive scheme is shown to guarantee the tracking error, between the outputs of system and the desired values, to be asymptotical in decay.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"9 36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129212798","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：