Abstract: A computational model of problem solving based on significant aspects of human problem solving is introduced. It is observed that during problem solving humans often start searching more or less randomly, becoming more deterministic over time as they learn more about the problem. This two-phase aspect of problem-solving behavior and its relation to learning is one of the important features this model accounts for. The model uses an accelerated simulated annealing technique as a search mechanism within a real-time dynamic programming-like framework upon a connected graph of neighboring problem states. The objective value of each node is adjusted as the model moves between nodes, learning more accurate values for the nodes and also compensating for misleading heuristic information as it does so. In this manner the model is shown to learn to more effectively solve isomorphs of the Balls and Boxes and Tower of Hanoi problems. The major issues investigated with the model are (a) whether such a simulated annealing-based model exhibits the kind of random-to-directed transition in behavior exhibited by people, and (b) whether the progressive discovery of the objective function, even when given very little or poor initial information, is a plausible method for representing the learning that occurs during problem solving and the knowledge that results from that learning.

Abstract: In this paper, neural networks trained with the back-propagation algorithm are applied to predict the future values of time series that consist of the weekly demand on items in a supermarket. The influencing indicators of prices, advertising campaigns and holidays are taken into consideration. The design and implementation of a neural network forecasting system is described that has been installed as a prototype in the headquarters of a German supermarket company to support the management in the process of determining the expected sale figures. The performance of the networks is evaluated by comparing them to two prediction techniques used in the supermarket now. The comparison shows that neural nets outperform the conventional techniques with regard to the prediction quality.

Abstract: The folding architecture is a universal mechanism to approximate mappings between trees and real vector spaces with a neural network. The part encoding the trees and the part approximating the mapping are trained simultaneously so that the encoding ts to the speciic learning task. In this article we show that this architecture is capable of approximating any mapping arbitrary well.

Abstract: In most of the practical applications the concept of vague environment [1] gives a simple way for fuzzy approximate reasoning. If the fuzzy partitions (used as primary sets of the fuzzy rulebase) can be described by vague environments [1], the primary fuzzy sets of the antecedent and the consequent parts of the fuzzy rules can be characterised by points in their vague environments. So the fuzzy rules themselves can be characterised by points in their vague environment too. It means, that the question of approximate fuzzy reasoning can be reduced to the problem of interpolation of the rule points in the vague environment of the fuzzy rulebase relation [2,3]. In this paper an approximate fuzzy reasoning method based on rational interpolation in the vague environment of the fuzzy rulebase will be introduced, and as an example of a practical application of the method, a path tracking control strategy for differential steered AGVs (Automated Guided Vehicle) [4] implemented on such a fuzzy logic controller will be introduced.

Abstract: Presuppositionis a pervasive feature of human language. It involves many interesting interactions between the utterances of a discourse and the contextof the discourse. In this paper we focus on issues of logical form connected with the interaction of presupposition and discourse context, and illustrate our theory with some implementational work using the active logicframework. After reviewing some of the major issues in presupposition theory we turn to a largely successful unified approach of Heim. We show how the main principles of this theory can be implemented in active logic. But we also find two serious difficulties. These consist in (a) a straightforward counterexample and (b) a type of discourse that we call a garden-path discourse. We maintain that both the counterexample and the garden-path type of discourse can be handled by our active-logic version of Heim's theory. This requires us to reformulate and extend Heim's theorey. Although this work is largely theoretical, both Heim's theory and ours have important things to say about the incremental processing of the utterances that make up discourse. And we present our theory as a specification of a processing device that takes logical form of a sentence along with current discourse context as input and delivers an updated discourse context as output. As an experiment, we have implemented portions of this device.

Abstract: In ordinary first–order logic, a valid inference in a language L is one in which the conclusion is true in every model of the language in which the premises are true. To accommodate inductive/uncertain/probabilistic/nonmonotonic inference, we weaken that demand to the demand that the conclusion be true in a large proportion of the models in which the relevant premises are true. More generally, we say that an inference is [p,q] valid if its conclusion is true in a proportion lying between p and q of those models in which the relevant premises are true. If we include a statistical variable binding operator “%” in our language, there are many quite general (and useful) things we can say about uncertain validity. A surprising result is that some of these things may conflict with Bayesian conditionalization.

Abstract: In description logics, default knowledge is exclusively treated as incidental rules. However, as few concepts are definable using only strict knowledge, imposing strict definitions leads to terminological knowledge bases that mostly contain partially defined concepts. This is a real problem because such concepts can only be inserted as leaves of the terminology. Moreover, instance recognition is biased as these concepts must be explicitly mentioned as properties of these instances. It follows that partially defined concepts are described with necessary but not sufficient conditions. As a solution to these problems, we propose to integrate defaults in concept definitions and we argue that this is essential for our diagnosis application. We introduce a description language ALeδ with default(δ) and exception(e) connectives. The cornerstone of our approach is the introduction of a definitional point of view where a default can be part of a concept definition, whereas in the classical inheritance one it is only viewed as a weak implication. We go on to describe a map between the definition of a concept and its inherited properties, and we show that the combination of these definitional and inheritance levels considerably improves the capabilities of classification processes. In particular this allows us to distinguish sure from probable instances and typical from exceptional instances. Finally we provide a specific operation, object refinement, which consists in enlarging object descriptions with exceptions in order to find additional concepts the object is an instance of. This operation is useful for our diagnosis application.

Abstract: An algorithm for semantic interpretation that integrates the determination of the meaning of verbs, the attachment and meaning of prepositions, and the determination of thematic roles is presented. The parser does not resolve structural ambiguity, which is solely the task of the semantic interpreter. Lexical semantic information about nouns and verbs is applied to the resolution of verb polysemy and modifier attachment. Semantic interpretation is centered on the representation of the meaning of the verb, called verbal concept. Verbal concepts are organized into a classification hierarchy. As long as the meaning of the verb remains unknown, parsing proceeds on a syntactic basis. Once the meaning of the verb is recognized, the semantic component makes sense of the syntactic relations built so far by the parser and of those still to be parsed. The algorithm has been implemented and tested on real–world texts.

Abstract: Several software component identification problems require evaluation of the fitness of a candidate on the basis of the information attached to it by a classification model. Fuzzy query algebras can effectively deal with these problems, choosing query execution mechanisms on the basis of the semantics selected by the user.

Abstract: Processor of new generation, so called fuzzy microprocessor, is used to control the position of header with infrarerd LED's. In its operations the processor use fuzzy logic rules and approximate reasoning. Hole system can be consider as intelligent infrared sensor, or system which track the infrared source and turn the header in the source direction. If it is necessary, intensity of infrared lights can be measured. In the paper design method of the system is described.

Abstract: Fifteen years of work on nonmonotonic logic has certainly increased our understanding of the area. However, given a problem in which nonmonotonic reasoning is called for, it is far from clear how one should go about modeling the problem using the various approaches. We explore this issue in the context on two of the best–known approaches, Reiter's default logic and Moore's autoepistemic logic, as well as two related notions of “only knowing,” due to Halpern and Moses and to Levesque. In particular, we return to the original technical definitions given in these papers and examine the extent to which they capture the intuitions they were designed to capture.

Abstract: A new fuzzy model structure identification method, based on orthogonalisation and statistical tests, as well as information criteria to obtain a minimum rule base and a minimum number of membership functions from input-output data, is proposed. The method is applied to functional-type fuzzy models. The applicability of the proposed method to nonlinear static and dynamic systems is illustrated by examples.

Abstract: Evolutionary algorithms have been studied for over 35 years. This paper provides a brief summary of the similarities and differences of various methods in evolutionary computation, as well as some ideas for future avenues of research.

Abstract: The development and usage of soft computing systems for forecasting of water level progress in case of flood events at river Mosel are presented. The practical situation and its requirements are explained and two different system approaches are discussed: a) a neural network for supervised learning of the functional behavior of time series data and its approximation, and b) a fuzzy system for modeling of the system behavior with possibilities to exploit expert information and for systematic optimization. Advantages and disadvantages of both concepts are described and emphasis is laid on the structural development of the fuzzy system. Both systems have been tested and satisfying results are shown with practical data.

Abstract: A representation scheme for verbs and prepositions specifying path and locative information is developed. The representation emphasizes the implementability of the underlying semantic primitives. The primitives pertain to mechanical characteristics such as geometric relationships among objects, force or motion characteristics implied by verbs, and their prepositional modifiers. This representation has been used to animate the performance of tasks underlying natural language imperatives.

Abstract: Kernel based neural networks with probabilistic reasoning are suitable for many practical applications. But influence of data set sizes let the probabilistic approach fail in case of small data amounts. Possibilistic reasoning avoids this drawback because it is independent of class size.

Abstract: We present a mathematical model providing a formal basis for analogical reasoning, referred to as abstractional concept mapping. The approach provides a formal basis for all comparison–based reasoning (e.g., literal similarity, analogy, metaphor, and scientific models) in that given any two descriptions in first–order logic, arbitrary comparisons of similarity are representable as first–order theories (referred to as “comparison theories”), dependent only on the notion of logical truth and not on domain–specific heuristics or particular features of the knowledge representation.

Abstract: In this paper the idea is presented that human communication is carried through messages that are the result of the integration of a set of communicative acts, some of which are, in a traditional view, contextual phenomena with respect to a main (linguistic) communication act. A formalization of the notion of communicative situation is attempted, which eliminates the distinction between context and main communication acts. The ideas presented in this paper come from a re-thinking of a prototype implemented in 1989 for the ESPRIT Project 527, CFID whose main features are presented in Section 4.

Abstract: There are numerous logical formalisms capable of drawing conclusions using default rules. Such systems, however, do not normally determine where the default rules come from; i.e., what it is that makes Birds flya good rule, but Birds drive trucksa bad one. Generic sentences such as Birds fly are often used informally to describe default rules. I propose to take this characterization seriously, and claim that a default rule is adequate if the corresponding generic sentence is true. Thus, if we know that Tweety is a bird, we may conclude by default that Tweety flies, just in case Birds fly is a true sentence. In this paper, a quantificational account of the semantics of generic sentences is presented. It is argued that a generic sentence is evaluated not in isolation, but with respect to a set of relevant alternatives. For example, Mammals bear live young is true because among mammals that bear live young, lay eggs, undergo mitosis, or engage in some alternative form of procreation, the majority bear live young. Since male mammals do not procreate in any form, they do not count. Some properties of alternatives are presented, and their interactions with the phenomena of focus and presupposition is investigated. It is shown how this account of generics can be used to characterize adequate default reasoning systems, and several desirable properties of such systems are proved. The problems of the automatic acquisition of rules from natural language are discussed. Because rules are often explicitly expressed as generics, it is argued that the interpretation of generic sentences plays a crucial role in this endeavor, and it is shown how the theory presented here can facilitate such interpretation.

Abstract: Intensional negative adjectives alleged, artificial, fake, false, former, and toyare unusual adjectives that depending on context may or may not be restricting functions. A formal theory of their semantics, pragmatics, and context that uniformly accounts for their complex mathematical and computational characteristics and captures some peculiarities of individual adjectives is presented. Such adjectives are formalized as new concept builders, negation-like functions that operate on the values of intensional properties of the concepts denoted by their arguments and yield new concepts whose intensional properties have values consistent with the negation of the old values. Understanding these new concepts involves semantics, pragmatics and context-dependency of natural language. It is argued that intensional negative adjectives can be viewed as a special-purpose, weaker, conntext-dependent negationin natural language. The theory explains and predicts many inferences licensed by expressions involving such adjectives. Implementation of sample examples demonstrates its computational feasibility. Computation of context-dependent interpretation is discussed. The theory allows one to enhance a knowledge representation system with similar concept building, negation-like, context-dependent functions, the availability of which appears to be a distinct characteristic of natural languages.

Abstract: The binary version of the school timetabling (STT) problem is a real-world example of a constraint network that includes only constraints of inequality. A new and useful representation for this real-world problem, the STT_Grid, leads to a generic decomposition technique. The paper presents proofs of necessary and sufficient conditions for the existence of a solution to decomposed STT_Grids. The decomposition procedure is of low enough complexity to be practical for large problems, such as a real-world high school. To test the decomposition approach, a typical high school was analyzed and used as a model for generating STT_Grids of various sizes. Experiments were conducted to test the difficulty of large STT networks and their solution by decomposition. The experimental results show that the decomposition procedure enables the solution of large STT_Grids (620 variables for a real school) in reasonable time. The constraint network of a typical STT_Grid is sparse and belongs to the class of easy problems. Still, due to the sizes of STTs, good constraint satisfaction problem search techniques (i.e., BackJumping and ForwardChecking) do not terminate in reasonable times for STT_Grids that are larger than 300 variables.

Abstract: A selforganized map was designed to learn and detect sleep stages will be described. Initial the input data were preprocessed with Difference Power Spectra (DPS (german: DLS)). The associative fields of the Kohonen map are directly transformed by frequency spectra. Interference phenomena are probably indicating the influence of biosignals with a source in the reticular system of a human brain.

Abstract: The aim of this paper is to provide a new method to compensate remaining unbalances at magnetic bearings. The compensating signal will be generated by a MLP-network. The main advantage that this new method presents lies in the simplified manner of operation rather than in a qualitative improvement of disturbance compensation.

Abstract: Recently, several significant advances have been made in two distinct theoretical areas. These theoretical advances have created an innovative field of theoretical and applied interest: fuzzy neural systems. Researchers have provided a theoretical basis in the field while industry has used this theoretical basis to create a new class of machines using the innovative technology of fuzzy neural networks. The theory of fuzzy logic provides a mathematical framework for capturing the uncertainties associated with human cognitive processes, such as thinking and reasoning. It also provides a mathematical morphology for emulating certain perceptual and linguistic attributes associated with human cognition. On the other hand, computational neural network paradigms have evolved in the process of understanding the incredible learning and adaptive features of neuronal mechanisms inherent in certain biological species. The integration of these two fields, fuzzy logic and neural networks, has the potential for combining the benefits of these two fascinating fields into a single capsule. The intent of this paper is to describe the basic notions of biological and computational neuronal morphologies, and to describe the principles and architectures of fuzzy neural networks.

Abstract: The standard AI method for representing temporal information in a first-order logic is to directly associate the information with a time point or interval via a relation. We present a new paradigm for temporal knowledge representation. All point-based temporal information is translated to real-valued functions in the Cartesian plane. For example, information that is true/false at a point becomes a 0–1 function. Other types of information, such as velocity, are directly represented with real-valued functions. The unique feature of the proposed approach is the use of the Riemann integral to represent interval-based information. Our approach is based on the fact that what is true at every point in an interval completely determines what is true over the interval. We conclude with a formal presentation of a first-order logic that is based on the proposed representation.

Abstract: Adaption of parameters during neural network training according to the actual shape of the error surface is supposed to be a powerful instrument to enforce convergence and to decrease time consumption of neural network training.

Abstract: A new fuzzy controller has been designed and compared with the one developed previously for the control of fed-batch yeast fermentation. The respiratory quotient (RQ) was used as controller input and flow rate of glucose solution was controlled to maximize conversion of glucose to biomass. The new controller was found to be stable throughout the production period, keeping the glucose flow rate at the desired value, whereas the old controller was unstable towards the end of the fermentation and was unable to control and minimize the ethanol concentration in the fermentation broth.

Abstract: This paper describes research undertaken while developing an interactive game based on a children’s science fiction film. The aim of the project is to develop artificial intelligence agents as characters in the game world which interact with the player around various predetermined scenarios. We describe the development of a representation for agent attributes, and how they affect on-going moods and changing relationships between agents during the game.