A area de revisão de crenças estuda como agentes racionais mudam suas crencas ao receberem novas informações. O marco da area de revisão de crenças foi a publicacão do trabalho de Alchourron, Gardenfors e Makinson. Nesse trabalho conhecido como paradigma AGM foram denidos criterios de racionalidade para tipos de mudanca de crencas. Desde então, a área de revisão de crenças foi influenciada por diversas disciplinas como filosoa, computacão e direito. Paralelamente ao desenvolvimento da area de revisão de crenças, os últimos 20 anos foram marcados por um grande avanço no estudo das logicas de descrição. Tal avanço, impulsionado pelo desenvolvimento da web-semântica, levou a adoção de linguagens inspiradas em logicas de descrição (OWL) como padrão para se representar ontologias na web. Nessa tese tratamos do problema de aplicar a teoria da revisão de crenças a lógicas não clássicas e especialmente a logicas de descric~ao. Trabalhos recentes mostraram que o paradigma AGM e incompatvel com diversas logicas de descricão. Estendemos esses resultados mostrando outras lógicas que não são compatíveis com o paradigma AGM. Propomos formas de aplicar a teoria de revisão tanto em bases quanto em conjuntos de crencas a essas logicas. Alem disso, usamos algoritmos conhecidos da área de depuração de ontologias para implementar operações em bases de crenças. / Belief revision theory studies how rational agents change their beliefs after receiving new information. The most in uential work in this area is the paper of Alchourron, Gardenfors and Makinson. In this work, known as AGM paradigm rationality criteria for belief change were dened. Since then, the eld has been in uenced by many areas like philosophy, computer science and law. Parallel to the development of belief revision eld, in the past 20 years there was a huge grow in the study of description logics. The climax of this development was the adoption of OWL (a language based on description logics) as the standard language to represent ontologies on the web. In this work we deal with the problem of applying belief revision in to non-classical logics, specially description logics. Recent works showed that the AGM paradigm is not compliant with several description logics. We have extended this work by showing that other logics are not compliant with AGM paradigm. Furthermore, we propose alternative ways to apply belief revision techniques to these logics. Finally, we show that well known algorithms from the area of ontology debugging eld can be used to implement the proposed constructions.

First, we consider different kinds of representation of uncertainty and the meth- ods for updating each of them by conditioning. We focus on the generic frame- work of (conditional) plausibility spaces, since it generalises all the introduced representations. Further, we select three frameworks and list the properties that need to be added to a conditional plausibility space in order to recover each of these frameworks. The main goal of this work, however, is to show how public announcement on single-agent plausibility models, ranking structures, and pos- sibility structures realised by their corresponding update mechanisms, can be embedded into the framework of conditional plausibility spaces. At the very end we briefly illustrate a general update model using plausibility measures. Keywords: belief revision, dynamic logic, epistemic logic, plausibility space, pub- lic announcement, uncertainty, update.

The ability to change one's beliefs in a rational manner is one of many facets of the abilities of an intelligent agent. Central to any investigation of belief change is the notion of an epistemic state. This dissertation is mainly concerned with three issues involving epistemic states: 1. How should an epistemic state be represented? 2. How does an agent use an epistemic state to perform belief change? 3. How does an agent arrive at a particular epistemic state? With regard to the first question, note that there are many different methods for constructing belief change operations. We argue that semantic constructions involving ordered pairs, each consisting of a set of beliefs and an ordering on the set of "possible worlds" (or equivalently, on the set of basic independent bits of information) are, in an important sense, more fundamental. Our answer to the second question provides indirect support for the use of semantic structures. We show how well-known belief change operations and related structures can be modelled semantically. Furthermore, we introduce new forms of belief change related operations and structures which are all defined, and motivated, in terms of such semantic representational formalisms. These include a framework for unifying belief revision and nonmonotonic reasoning, new versions of entrenchment orderings on beliefs, novel approaches to withdrawal operations, and an expanded view of iterated belief change. The third question is. one which has not received much attention in the belief change literature. We propose to extract extra-logical information from the formal representation of an agent's set of beliefs, which can then be used in the construction of epistemic state. his proposal is just a first approximation, although it seems to have the potential for developing into a full-fledged theory. / Computing / D.Phil.(Computer Science)

The inadequacy of the usual possible world semantics of modal languages when the meaning of 'belief' is attached to the modal operator is discussed. Three other approaches are then investigated. In the case of Moore's autoepistemic logic it becomes possible to compare an agent's beliefs to 'reality', which cannot be done directly in the possible world semantics. Levesque's semantics makes explicit in the object language the notion of 'this is all the information the agent has', which plays an important role in nonmonotonic reasoning. Both of these approaches deal with ideal reasoners. The third approach, Konolige's deduction model, is based on a semantics capable of describing the beliefs of one or more resourcebounded agents. Finally, the AGM postulates for belief revision are discussed. / Computer Science / M.Sc. (Computer Science)

Cette thèse s'inscrit dans le domaine de la représentation des connaissances et du raisonnement en Intelligence Artificielle. Elle traite divers aspects du changement de croyances dans le cadre de fragments de la logique propositionnelle.Dans un premier temps, nous nous intéressons à la complexité du problème de vérification de modèle pour des opérateurs de révision de bases de croyances dans le cadre général de la logique propositionnelle et dans le cadre restreint des formules de Horn et des formules de Krom.Notre contribution principale porte ensuite sur le raffinement des opérateurs de changement de croyances afin que ceux-ci opèrent dans des fragments de la logique propositionnelle. Nous examinons en particulier les opérations de révision, de mise-à-jour et de contraction. Cette approche permet, dans chacun des cas, d'obtenir des opérateurs concrets, dont nous étudions les propriétés logiques en terme de de satisfaction de postulats que doivent satisfaire les opérateurs de changement de croyances rationnels. Divers fragments de la logique propositionnelle sont considérés, notamment les fragment de Horn et de Krom. / This thesis takes place in the field of knowledge representation and reasoning in Artificial Intelligence.It deals with various issues of belief change within fragments of propositional logic.First we focus on the complexity of model-checking for different revision operators within the general framework of propositional logic and within the framework of Horn and Krom fragments.Second, our main contribution is the study of the refinement of belief change operators in such a way that they act within fragments of propositional logic. In particular, we address refinement of revision, update and contraction operators. In each case this approach allows us to define concrete operators, for which we study logical properties in terms of satisfaction of postulates that should hold for any rational belief change operator. Various propositional fragments of propositional logic are considered, such as Horn and Krom fragments.

Cette thèse propose un modèle permettant la mise en œuvre d'un système de raisonnement à partir de cas capable d'adapter des procédures représentées sous forme de texte en langue naturelle, en réponse à des requêtes d'utilisateurs. Bien que les cas et les solutions soient sous forme textuelle, l'adaptation elle-même est d'abord appliquée à un réseau de contraintes temporelles exprimées à l'aide d'une algèbre qualitative, grâce à l'utilisation d'un opérateur de révision des croyances. Des méthodes de traitement automatique des langues sont utilisées pour acquérir les représentations algébriques des cas ainsi que pour regénérer le texte à partir du résultat de l'adaptation / This thesis proposes a practical model making it possible to implement a case-based reasoning system that adapts processes represented as natural language text in response to user queries. While the cases and the solutions are in textual form, the adaptation itself is performed on networks of temporal constraints expressed with a qualitative algebra, using a belief revision operator. Natural language processing methods are used to acquire case representations and to regenerate text based on the adaptation result

The ability to change one's beliefs in a rational manner is one of many facets of the abilities of an intelligent agent. Central to any investigation of belief change is the notion of an epistemic state. This dissertation is mainly concerned with three issues involving epistemic states: 1. How should an epistemic state be represented? 2. How does an agent use an epistemic state to perform belief change? 3. How does an agent arrive at a particular epistemic state? With regard to the first question, note that there are many different methods for constructing belief change operations. We argue that semantic constructions involving ordered pairs, each consisting of a set of beliefs and an ordering on the set of "possible worlds" (or equivalently, on the set of basic independent bits of information) are, in an important sense, more fundamental. Our answer to the second question provides indirect support for the use of semantic structures. We show how well-known belief change operations and related structures can be modelled semantically. Furthermore, we introduce new forms of belief change related operations and structures which are all defined, and motivated, in terms of such semantic representational formalisms. These include a framework for unifying belief revision and nonmonotonic reasoning, new versions of entrenchment orderings on beliefs, novel approaches to withdrawal operations, and an expanded view of iterated belief change. The third question is. one which has not received much attention in the belief change literature. We propose to extract extra-logical information from the formal representation of an agent's set of beliefs, which can then be used in the construction of epistemic state. his proposal is just a first approximation, although it seems to have the potential for developing into a full-fledged theory. / Computing / D.Phil.(Computer Science)

The inadequacy of the usual possible world semantics of modal languages when the meaning of 'belief' is attached to the modal operator is discussed. Three other approaches are then investigated. In the case of Moore's autoepistemic logic it becomes possible to compare an agent's beliefs to 'reality', which cannot be done directly in the possible world semantics. Levesque's semantics makes explicit in the object language the notion of 'this is all the information the agent has', which plays an important role in nonmonotonic reasoning. Both of these approaches deal with ideal reasoners. The third approach, Konolige's deduction model, is based on a semantics capable of describing the beliefs of one or more resourcebounded agents. Finally, the AGM postulates for belief revision are discussed. / Computer Science / M.Sc. (Computer Science)