The need for negotiation in multi-agent systems stems from the requirement for agents to solve the problems posed by their interdependence upon one another. Negotiation provides a solution to these problems by giving the agents the means to resolve their conflicting objectives, correct inconsistencies in their knowledge of other agents' world view, and coordinate a joint approach to domain tasks which benefits all the agents concerned. We propose a framework, based upon a system of argumentation, which permits agents to negotiate to establish acceptable ways to solve problems.

Systems of argumentation for handling beliefs about the world have been reported in earlier papers. It seems possible that these systems may also be applicable to reasoning about the effects of actions. However there are substantial differences reasoning about beliefs and reasoning about actions, so a new system of argumentation is required for the latter. This paper makes some preliminary remarks about how the argumentation framework we have introduced elsewhere might be extended to making decisions about the expected value of actions. Introduction Standard decision theory (Raiffa 1970) builds on the probabilistic view of uncertainty in reasoning about actions. The costs and benefits of possible outcomes of actions are weighted with their probabilities, yielding a preference ordering on the "expected utility" of alternative actions. However, as Tan and Pearl (1994), amongst others, have pointed out, the specification of the complete sets of probabilities and utilities requ...

Numerous argumentation systems have been proposed in the literature. Yet there often appears to be a shortfall between proposed systems and possible applications. In other words, there seems to be a need for further development of proposals for argumentation systems before they can be used widely in decision-support or knowledge management. We believe that this shortfall can be bridged by taking a hybrid approach. Whilst formal foundations are vital, systems that incorporate some of the practical ideas found in some of the informal approaches may make the resulting hybrid systems more useful. In informal approaches, there is often an emphasis on using graphical notation with symbols that relate more closely to the real-world concepts to be modelled. There may also be the incorporation of an argument ontology oriented to the user domain. Furthermore, in informal approaches there can be greater consideration of how users interact with the models, such as allowing users to edit arguments and to weight influences on graphs representing arguments. In this paper, we discuss some of the features of argumentation, review some key formal argumentation systems, identify some of the strengths and weaknesses of these formal proposals, and finally consider some ways to develop formal proposals to give hybrid argumentation systems. To focus our discussions, we will consider some applications, in particular an application in analysing structured news reports.

. This paper summarises our position on the use of symbolic methods for reasoning under uncertainty, and argumentation in particular. Our view is that argumentation offers a complement to numerical methods for reasoning about belief, and a general framework within which many competing approaches can be understood. In applications we have found that argumentation offers a variety of benefits for practical reasoning systems. The presentation is historical, emphasising the reasons which motivated the development of the argumentation framework, drawing primarily on work carried out by researchers at the Imperial Cancer Research Fund since about 1980. 1 The need for symbolic decision making Work on argumentation at the Imperial Cancer Research Fund arose out of a series of studies on medical decision making, including work on modelling human diagnostic reasoning [9] and comparisons of the relative merits of numerical and symbolic inference techniques in clinical decision making [11, 16, 22...

This paper discusses a system of argumentation with a probabilistic semantics and compares it to two other probabilistic systems---Wellman's qualitative probabilistic networks and Neufeld's probabilistic default reasoning. 1 INTRODUCTION In the last few years there have been a number of attempts to build systems for reasoning under uncertainty that are of a qualitative nature---that is they use qualitative rather than numerical values, dealing with concepts such as increases in belief and the relative magnitude of values. In particular, two types of qualitative system have become well established--- qualitative probabilistic networks (QPNs) [2, 12], and systems of argumentation [5, 6]. While the former are built as an abstraction of probabilistic networks where the links between nodes are only modelled in terms of the qualitative influence of the parents on the children, and therefore have an underlying probabilistic semantics, some of the latter lack such a sound foundation. This lac...

To operate in the real-world, intelligent agents constantly need to absorb new information, and to consider the ramifications of it. This raises interesting questions for knowledge representation and reasoning. Here we consider ramification analysis in which we wish to determine both the likely outcomes from events occuring and the less likely, but very significant outcomes, from events occuring. To formalize ramification analysis, we introduce the notion of causal maps for modelling "causal relationships" between events. In particular, we consider existential event classes, for example presidential-election, with instances being true, false, or unknown, and directional events classes, for example inflation, with instances being increasing, decreasing or unchanging. Using causal maps, we can propagate new information to determine possible ramifications. These ramifications are also described in terms of events.

Weighted Factors Analysis (WeFA) has been proposed as a new approach for elicitation, representation, and manipulation of knowledge about a given problem, generally at a high and strategic level. Central to this proposal is that a group of experts in the area of the problem can identify a hierarchy of factors with positive or negative influences on the problem outcome.