Michio Sugeno - Academia.edu (original) (raw)
Papers by Michio Sugeno
Communications in Computer and Information Science, 2014
Anti-windup methods deal with saturation problems in systems with known actuator limits. Part of ... more Anti-windup methods deal with saturation problems in systems with known actuator limits. Part of the controller design is devoted to constraint handling in order to avoid pernicious behaviour of the systems. Feedback error learning is an on-line learning strategy of inverse dynamics. It sequentially acquires an inverse model of a plant through feedback control actions. In previous works we proposed an approach to implement the FEL control scheme through Piecewise Bilinear models. In this paper, a PB based FEL implementation scheme is proposed, where saturations provoked by the input constraints are taken into account for the inverse model learning algorithm.
FUZZ-IEEE'99. 1999 IEEE International Fuzzy Systems. Conference Proceedings (Cat. No.99CH36315)
The authors propose an extended subset interactive autoregressive (AR) model using non-monotonic ... more The authors propose an extended subset interactive autoregressive (AR) model using non-monotonic fuzzy measures and the Choquet integral. Genetic algorithms (GAs) and the statistical model selection criterion (i.e., Akaike's information criterion, AIC) are used for the identification of the proposed model. The fitting and forecasting performances of the proposed model are analyzed by applying it to well-known time series data, the sunspot series. The results show that the proposed model is useful for time series analysis with nonlinear features
IEEE SMC'99 Conference Proceedings. 1999 IEEE International Conference on Systems, Man, and Cybernetics (Cat. No.99CH37028)
Human activity naturally contains ambiguities and contradictions that cannot be solved mathematic... more Human activity naturally contains ambiguities and contradictions that cannot be solved mathematically or logically. We think that human intelligence is closely related to the use of language. Therefore, in order to achieve intelligent computing as human beings do, we need to realize the architecture in which language is used as an information medium on a computer. In this paper we study on the development of a travel consultation system based on the above architecture as an example of information processing in intelligent computing, and show how language functions in information processing
Fuzzy Approaches for Soft Computing and Approximate Reasoning: Theories and Applications
2004 IEEE International Conference on Systems, Man and Cybernetics (IEEE Cat. No.04CH37583)
ABSTRACT
Proceedings of 1995 IEEE International Conference on Fuzzy Systems. The International Joint Conference of the Fourth IEEE International Conference on Fuzzy Systems and The Second International Fuzzy Engineering Symposium
ABSTRACT In this paper, a new framework for interpreting linguistic labels in executing fuzzy pro... more ABSTRACT In this paper, a new framework for interpreting linguistic labels in executing fuzzy programs is proposed. The basic notion of a fuzzy program is that of a “linguistic program”, which is a human friendly programming system. The FAPS2 system, an experimental framework for executing fuzzy programs, is also proposed. In this paper, the authors discuss the issues of fuzzy programs through a practical example: a “GO” game playing system
Firstly in this paper we propose a new criterion function for evaluation of the reductive propert... more Firstly in this paper we propose a new criterion function for evaluation of the reductive property about the fuzzy reasoning result for fuzzy modus ponens and fuzzy modus tollens. Secondly unlike fuzzy reasoning methods based on the similarity measure, we propose a new fuzzy reasoning method based on distance measure. Thirdly the reductive property for 5 fuzzy reasoning methods are checked with respect to fuzzy modus ponens and fuzzy modus tollens. Through the experiment, we show that proposed method is better than the previous methods in accordance with human thinking.
IEEE SMC'99 Conference Proceedings. 1999 IEEE International Conference on Systems, Man, and Cybernetics (Cat. No.99CH37028)
A construction procedure of a quadratic Lyapunov function is proposed to analyze the stability of... more A construction procedure of a quadratic Lyapunov function is proposed to analyze the stability of fuzzy systems with singleton consequents. LMI (linear matrix inequality) and sequential quadratic programming are used to find the Lyapunov function which satisfies the stability conditions. The stability conditions are classified into two parts: conditions which can be converted to LMI form by introducing slack variables and by using Schur complements, and conditions which can not be converted. The proposed method is illustrated by a numerical example.
Advances in Fuzzy Sets, Possibility Theory, and Applications
We have heuristically designed fuzzy controllers so far since we have lacked a fuzzy model of a s... more We have heuristically designed fuzzy controllers so far since we have lacked a fuzzy model of a system. The authors have recently developed a method of multi-dimensional fuzzy reasoning that enables one to build a dynamic model of a system just as we do in terms of differential equations. This paper presents a new idea of designing a fuzzy controller based on a fuzzy model of a system.
Journal of Japan Society for Fuzzy Theory and Intelligent Informatics
Journal of Japan Society for Fuzzy Theory and Intelligent Informatics
Journal of Japan Society for Fuzzy Theory and Systems
IEEE Transactions on Fuzzy Systems
This paper deals with distorted statistics as statistics based on distorted probabilities by appl... more This paper deals with distorted statistics as statistics based on distorted probabilities by applying Choquet calculus. First, elementary distorted statistics on the nonnegative line is explored for exponential and Gaussian distributions. Next, by introducing Choquet calculus for real-valued functions on the real line, distorted statistics is explored for uniform and normal distributions. Finally, the relations of distorted statistics with conventional skew statistics are clarified.
Journal of Advanced Computational Intelligence and Intelligent Informatics
In this paper, we demonstrate the modeling capabilities of certain types of fuzzy integrals such ... more In this paper, we demonstrate the modeling capabilities of certain types of fuzzy integrals such as hierarchical bipolar/cumulative-prospect-theory-type Sugeno and Choquet integrals. The notion of admissible preference structures, introduced by Nakama and Sugeno, is one of the weakest restrictions in rational preference structures. Nakama and Sugeno also proved that, under a certain condition, any admissible preference can be modeled by a hierarchical bipolar Sugeno integral. Here, we extend this result and show that if we use an extra dummy attribute, we can use the hierarchical Choquet integral to generate any admissible preference.
Journal of Advanced Computational Intelligence and Intelligent Informatics
In order to develop a brain machine interface, the authors have investigated the brain activity d... more In order to develop a brain machine interface, the authors have investigated the brain activity during human recognition of characters and symbols representing directional meaning. They have recorded electroencephalograms (EEGs) from subjects viewing four types of Kanji (Chinese characters being used currently in the Japanese language) and arrows that were presented on a CRT. Each of the four characters or symbols denoted direction for upward, downward, leftward and rightward, respectively. Subjects were asked to read the characters or symbols, silently. EEGs were averaged for each stimulus type and direction, and event related potentials (ERPs) were obtained. The equivalent current dipole source localization (ECDL) method has been applied to these ERPs. In both cases, equivalent current dipoles (ECDs) were localized to areas related to the working memory for spatial perception, such as the right upper or the right middle frontal areas. Taking into account these facts, the authors h...
Journal of Advanced Computational Intelligence and Intelligent Informatics
We propose a computer communication protocol based on natural language called "language prot... more We propose a computer communication protocol based on natural language called "language protocol", communication using the protocol, and an interface enabling connection any communication standard, called a "language application programming interface". We use simulation to confirm that the proposed methods provide a flexible communication environment for any communication object.
Journal of Advanced Computational Intelligence and Intelligent Informatics
We recorded event-related potentials (ERPs) by electroencephalography (EEG) during fuzzy and cris... more We recorded event-related potentials (ERPs) by electroencephalography (EEG) during fuzzy and crisp calculation. Questions art divided into 2 types. As type A, questions were presented as sentences. Questions of type B were presented as numerical calculation. In type A, the peak latency of the EEG was around 1100ms. In type B, the peak latency was around 650ms. In type A, from multiple equivalent current dipole source localization (ECDL) around the latency, it followed that sources during fuzzy calculation lie in the right hemisphere and that sources during crisp calculation lie in the left hemisphere. In type B, no significant difference was observed.
Journal of Advanced Computational Intelligence and Intelligent Informatics
The smart help we propose based on application software manuals features four phases: (1) underst... more The smart help we propose based on application software manuals features four phases: (1) understanding the user’s input text, which is a question about software operation; (2) matching the result of understanding with software manual text; (3) planning a dialog with the user; and (4) paraphrasing matching manual text to output for the user. Manual text found during matching is paraphrased using the analysis result found during understanding. In planning, the rhetorical structure of the selected manual is used to determine which clauses in matching manual text are specifically relevant to the user’s question. Four paraphrases for a novice user are identified in a subject experiment, and rules for these are applied based on linguistic features extracted during the understanding. We evaluated an implemented prototype, and discuss its applications and problems.
Journal of Advanced Computational Intelligence and Intelligent Informatics
This paper explains an overview of everyday language computing (ELC) promoted by the Laboratory f... more This paper explains an overview of everyday language computing (ELC) promoted by the Laboratory for Language-Based Intelligent Systems, Brain Science Institute, RIKEN from 2000 to 2005. The objective of ELC was to develop a language-based intelligent system. To do this, we constructed a computational model of language in context, called the Semiotic Base, and developed computational algorithms for text understanding and generation, which are basic information processing in the ELC framework. Based on these resources and algorithms, we constructed a computing environment in which language is used as an information medium to process information. To demonstrate its feasibility, we developed language-based applications such as a language-based wordprocessor, language-based programming, and “smart” help. We explain the basic principles of ELC, outline of its basic technologies, and discuss applications developed based on the ELC framework.
Journal of Advanced Computational Intelligence and Intelligent Informatics
We propose two computational models - one of a language within context based on systemic function... more We propose two computational models - one of a language within context based on systemic functional linguistic theory and one of context-sensitive language understanding. The model of a language within context called the Semiotic Base characterizes contextual, semantic, lexicogrammatical, and graphological aspects of input texts. The understanding process is divided into shallow and deep analyses. Shallow analysis consists of morphological and dependency analyses and word concept and case relation assignment, mainly by existing natural language processing tools and machine-readable dictionaries. Results are used to detect the contextual configuration of input text in contextual analysis. This is followed by deep analyses of lexicogrammar, semantics, and concepts, conducted by referencing a subset of resources related to the detected context. Our proposed models have been implemented in Java and verified by integrating them into such applications as dialog-based question-and-answer (...
Communications in Computer and Information Science, 2014
Anti-windup methods deal with saturation problems in systems with known actuator limits. Part of ... more Anti-windup methods deal with saturation problems in systems with known actuator limits. Part of the controller design is devoted to constraint handling in order to avoid pernicious behaviour of the systems. Feedback error learning is an on-line learning strategy of inverse dynamics. It sequentially acquires an inverse model of a plant through feedback control actions. In previous works we proposed an approach to implement the FEL control scheme through Piecewise Bilinear models. In this paper, a PB based FEL implementation scheme is proposed, where saturations provoked by the input constraints are taken into account for the inverse model learning algorithm.
FUZZ-IEEE'99. 1999 IEEE International Fuzzy Systems. Conference Proceedings (Cat. No.99CH36315)
The authors propose an extended subset interactive autoregressive (AR) model using non-monotonic ... more The authors propose an extended subset interactive autoregressive (AR) model using non-monotonic fuzzy measures and the Choquet integral. Genetic algorithms (GAs) and the statistical model selection criterion (i.e., Akaike's information criterion, AIC) are used for the identification of the proposed model. The fitting and forecasting performances of the proposed model are analyzed by applying it to well-known time series data, the sunspot series. The results show that the proposed model is useful for time series analysis with nonlinear features
IEEE SMC'99 Conference Proceedings. 1999 IEEE International Conference on Systems, Man, and Cybernetics (Cat. No.99CH37028)
Human activity naturally contains ambiguities and contradictions that cannot be solved mathematic... more Human activity naturally contains ambiguities and contradictions that cannot be solved mathematically or logically. We think that human intelligence is closely related to the use of language. Therefore, in order to achieve intelligent computing as human beings do, we need to realize the architecture in which language is used as an information medium on a computer. In this paper we study on the development of a travel consultation system based on the above architecture as an example of information processing in intelligent computing, and show how language functions in information processing
Fuzzy Approaches for Soft Computing and Approximate Reasoning: Theories and Applications
2004 IEEE International Conference on Systems, Man and Cybernetics (IEEE Cat. No.04CH37583)
ABSTRACT
Proceedings of 1995 IEEE International Conference on Fuzzy Systems. The International Joint Conference of the Fourth IEEE International Conference on Fuzzy Systems and The Second International Fuzzy Engineering Symposium
ABSTRACT In this paper, a new framework for interpreting linguistic labels in executing fuzzy pro... more ABSTRACT In this paper, a new framework for interpreting linguistic labels in executing fuzzy programs is proposed. The basic notion of a fuzzy program is that of a “linguistic program”, which is a human friendly programming system. The FAPS2 system, an experimental framework for executing fuzzy programs, is also proposed. In this paper, the authors discuss the issues of fuzzy programs through a practical example: a “GO” game playing system
Firstly in this paper we propose a new criterion function for evaluation of the reductive propert... more Firstly in this paper we propose a new criterion function for evaluation of the reductive property about the fuzzy reasoning result for fuzzy modus ponens and fuzzy modus tollens. Secondly unlike fuzzy reasoning methods based on the similarity measure, we propose a new fuzzy reasoning method based on distance measure. Thirdly the reductive property for 5 fuzzy reasoning methods are checked with respect to fuzzy modus ponens and fuzzy modus tollens. Through the experiment, we show that proposed method is better than the previous methods in accordance with human thinking.
IEEE SMC'99 Conference Proceedings. 1999 IEEE International Conference on Systems, Man, and Cybernetics (Cat. No.99CH37028)
A construction procedure of a quadratic Lyapunov function is proposed to analyze the stability of... more A construction procedure of a quadratic Lyapunov function is proposed to analyze the stability of fuzzy systems with singleton consequents. LMI (linear matrix inequality) and sequential quadratic programming are used to find the Lyapunov function which satisfies the stability conditions. The stability conditions are classified into two parts: conditions which can be converted to LMI form by introducing slack variables and by using Schur complements, and conditions which can not be converted. The proposed method is illustrated by a numerical example.
Advances in Fuzzy Sets, Possibility Theory, and Applications
We have heuristically designed fuzzy controllers so far since we have lacked a fuzzy model of a s... more We have heuristically designed fuzzy controllers so far since we have lacked a fuzzy model of a system. The authors have recently developed a method of multi-dimensional fuzzy reasoning that enables one to build a dynamic model of a system just as we do in terms of differential equations. This paper presents a new idea of designing a fuzzy controller based on a fuzzy model of a system.
Journal of Japan Society for Fuzzy Theory and Intelligent Informatics
Journal of Japan Society for Fuzzy Theory and Intelligent Informatics
Journal of Japan Society for Fuzzy Theory and Systems
IEEE Transactions on Fuzzy Systems
This paper deals with distorted statistics as statistics based on distorted probabilities by appl... more This paper deals with distorted statistics as statistics based on distorted probabilities by applying Choquet calculus. First, elementary distorted statistics on the nonnegative line is explored for exponential and Gaussian distributions. Next, by introducing Choquet calculus for real-valued functions on the real line, distorted statistics is explored for uniform and normal distributions. Finally, the relations of distorted statistics with conventional skew statistics are clarified.
Journal of Advanced Computational Intelligence and Intelligent Informatics
In this paper, we demonstrate the modeling capabilities of certain types of fuzzy integrals such ... more In this paper, we demonstrate the modeling capabilities of certain types of fuzzy integrals such as hierarchical bipolar/cumulative-prospect-theory-type Sugeno and Choquet integrals. The notion of admissible preference structures, introduced by Nakama and Sugeno, is one of the weakest restrictions in rational preference structures. Nakama and Sugeno also proved that, under a certain condition, any admissible preference can be modeled by a hierarchical bipolar Sugeno integral. Here, we extend this result and show that if we use an extra dummy attribute, we can use the hierarchical Choquet integral to generate any admissible preference.
Journal of Advanced Computational Intelligence and Intelligent Informatics
In order to develop a brain machine interface, the authors have investigated the brain activity d... more In order to develop a brain machine interface, the authors have investigated the brain activity during human recognition of characters and symbols representing directional meaning. They have recorded electroencephalograms (EEGs) from subjects viewing four types of Kanji (Chinese characters being used currently in the Japanese language) and arrows that were presented on a CRT. Each of the four characters or symbols denoted direction for upward, downward, leftward and rightward, respectively. Subjects were asked to read the characters or symbols, silently. EEGs were averaged for each stimulus type and direction, and event related potentials (ERPs) were obtained. The equivalent current dipole source localization (ECDL) method has been applied to these ERPs. In both cases, equivalent current dipoles (ECDs) were localized to areas related to the working memory for spatial perception, such as the right upper or the right middle frontal areas. Taking into account these facts, the authors h...
Journal of Advanced Computational Intelligence and Intelligent Informatics
We propose a computer communication protocol based on natural language called "language prot... more We propose a computer communication protocol based on natural language called "language protocol", communication using the protocol, and an interface enabling connection any communication standard, called a "language application programming interface". We use simulation to confirm that the proposed methods provide a flexible communication environment for any communication object.
Journal of Advanced Computational Intelligence and Intelligent Informatics
We recorded event-related potentials (ERPs) by electroencephalography (EEG) during fuzzy and cris... more We recorded event-related potentials (ERPs) by electroencephalography (EEG) during fuzzy and crisp calculation. Questions art divided into 2 types. As type A, questions were presented as sentences. Questions of type B were presented as numerical calculation. In type A, the peak latency of the EEG was around 1100ms. In type B, the peak latency was around 650ms. In type A, from multiple equivalent current dipole source localization (ECDL) around the latency, it followed that sources during fuzzy calculation lie in the right hemisphere and that sources during crisp calculation lie in the left hemisphere. In type B, no significant difference was observed.
Journal of Advanced Computational Intelligence and Intelligent Informatics
The smart help we propose based on application software manuals features four phases: (1) underst... more The smart help we propose based on application software manuals features four phases: (1) understanding the user’s input text, which is a question about software operation; (2) matching the result of understanding with software manual text; (3) planning a dialog with the user; and (4) paraphrasing matching manual text to output for the user. Manual text found during matching is paraphrased using the analysis result found during understanding. In planning, the rhetorical structure of the selected manual is used to determine which clauses in matching manual text are specifically relevant to the user’s question. Four paraphrases for a novice user are identified in a subject experiment, and rules for these are applied based on linguistic features extracted during the understanding. We evaluated an implemented prototype, and discuss its applications and problems.
Journal of Advanced Computational Intelligence and Intelligent Informatics
This paper explains an overview of everyday language computing (ELC) promoted by the Laboratory f... more This paper explains an overview of everyday language computing (ELC) promoted by the Laboratory for Language-Based Intelligent Systems, Brain Science Institute, RIKEN from 2000 to 2005. The objective of ELC was to develop a language-based intelligent system. To do this, we constructed a computational model of language in context, called the Semiotic Base, and developed computational algorithms for text understanding and generation, which are basic information processing in the ELC framework. Based on these resources and algorithms, we constructed a computing environment in which language is used as an information medium to process information. To demonstrate its feasibility, we developed language-based applications such as a language-based wordprocessor, language-based programming, and “smart” help. We explain the basic principles of ELC, outline of its basic technologies, and discuss applications developed based on the ELC framework.
Journal of Advanced Computational Intelligence and Intelligent Informatics
We propose two computational models - one of a language within context based on systemic function... more We propose two computational models - one of a language within context based on systemic functional linguistic theory and one of context-sensitive language understanding. The model of a language within context called the Semiotic Base characterizes contextual, semantic, lexicogrammatical, and graphological aspects of input texts. The understanding process is divided into shallow and deep analyses. Shallow analysis consists of morphological and dependency analyses and word concept and case relation assignment, mainly by existing natural language processing tools and machine-readable dictionaries. Results are used to detect the contextual configuration of input text in contextual analysis. This is followed by deep analyses of lexicogrammar, semantics, and concepts, conducted by referencing a subset of resources related to the detected context. Our proposed models have been implemented in Java and verified by integrating them into such applications as dialog-based question-and-answer (...