this paper, we present a solution to the problem how a team of robotic agents can come up with a hypothesis of what might globally be wrong with the environment in similar situations. Hypotheses should entail the currently observed behavior and provide some kind of "forecast" for those areas that no other member of the team visited so far. As time proceeds the hypothesis will be refined to match the actual situation more closely. With a hypothesis about the condition of the environment, single robots can try to avoid areas that potentially contain obstacles, take the shortest way out of these areas, or instead enter these areas to (dis-)prove the hypothesis dependent on the strategy of the team

Various formal theories have been proposed in the literature to capture the notions of diagnosis underlying diagnostic programs. Examples of such notions are: heuristic classification, which is used in systems incorporating empirical knowledge, and model-based diagnosis, which is used in diagnostic systems based on detailed domain models. Typically, such domain models include knowledge of causal, structural, and functional interactions among modelled objects. In this paper, a new set-theoretical framework for the analysis of diagnosis is presented. Basically, the framework distinguishes between `evidence functions', which characterize the net impact of knowledge bases for purposes of diagnosis, and `notions of diagnosis', which define how evidence functions are to be used to map findings observed for a problem case to diagnostic solutions. This set-theoretical framework offers a simple, yet powerful tool for comparing existing notions of diagnosis, as well as for proposing new notions ...

Sense-making software for crime investigation should be based on a model of reasoning about evidence that is both natural and rationally well-founded. A formal model is proposed that combines AI formalisms for abductive inference to the best explanation and for defeasible argumentation. Stories about what might have happened in a case are represented as causal networks and possible hypotheses can be inferred by abductive reasoning. Links between stories and the available evidence are expressed with evidential generalisations that express how observations can be inferred from evidential sources with defeasible argumentation. It is argued that this approach unifies two well-known accounts of reasoning about evidence, namely, anchored narratives theory and new evidence theory. After the reasoning model is defined, a design is presented for sense-making software that allows crime investigators to visualise their thinking about a case in terms of the reasoning model.

Model-based diagnosis concerns using a model of the structure and behaviour of a system or device in order to establish why the system or device is malfunctioning. Traditionally, little attention has been given to the problem of dealing with uncertainty in model-based diagnosis. Given the fact that determining a diagnosis for a problem almost always involves uncertainty, this situation is not entirely satisfactory. This paper builds upon and extends previous work in model-based diagnosis by supplementing the well-known model-based framework with mathematically sound ways for dealing with uncertainty.

f.j.bex at rug.nl In the present paper, we provide a formalised version of a merged argumentative and story-based approach towards the analysis of evidence. As an application, we are able to show how our approach sheds new light on inference to the best explanation with case evidence. More specifically, it will be clarified how the events in a case story that are considered to be proven abductively explain the otherwise unproven events of the case story. We compare our approach with existing AI work on modelling legal reasoning with evidence.

This paper addresses the design of an ambient agent model that incorporates model-based reasoning methods for the analysis of internal causes of observed undesired behaviours of a human, and for determination of actions that remedy such causes. The models used are based on causal and dynamical relations and integrate numerical aspects. By the model-based reasoning methods hypotheses, observations and actions are generated. Control parameters within these processes are described that allow the ambient agent to focus the reasoning. These control parameters are related to each other and to specific domain and situation characteristics, such as time pressure, or criticality of a situation. 1.

Abstract In this paper we propose a formal dialogue game in which two players aim to determine the best explanation for a set of observations. By assuming an adversarial setting, we force the players to advance and improve their own explanations as well as criticize their opponent’s explanations, thus hopefully

Almost two decades after the introduction of probabilistic expert systems, their theoretical status, practical use, and experiences are matching those of rule-based expert systems. Since both types of systems are in wide use, it is more than ever important to understand their advantages and drawbacks. We describe a study in which we compare rule-based systems to systems based on Bayesian networks. We present two expert systems for diagnosis of liver disorders that served as the inspiration and vehicle of our study and discuss problems related to knowledge engineering using the two approaches. We finally present the results of a simple experiment comparing the diagnostic performance of each of the systems on a subset of their domain.

Abstract not found

We present a prototype of the AICronus system, an argumentation system that automates a challenging reasoning process used by experts in cosmogenic isotope dating. The architecture of the system is described and preliminary results are discussed. 1.