In the literature on reasoning on the basis of evidence, two traditions exist: one argument-based, and one based on narratives. Recently, we have proposed a hybrid perspective in which argumentation and narratives are combined. This formalized hybrid theory has been tested in a sense-making software prototype for criminal investigators and decision makers. In the present paper, we elaborate on the role of commonsense knowledge. We argue that two kinds of knowledge are essential: argumentation schemes and story schemes. We discuss some of the research issues that need to be addressed.

Human experts in scientific fields routinely work with evidence that is noisy and untrustworthy, heuristics that are unproven, and possible conclusions that are contradictory. We present a fully implemented AI system, Calvin, for cosmogenic isotope dating, a domain that is fraught with these difficult issues. Calvin solves these problems using an argumentation framework and a system of confidence that uses two-dimensional vectors to express the quality of heuristics and the applicability of evidence. The arguments it produces are strikingly similar to published expert arguments. Calvin is in daily use by isotope dating experts.

Tournament solution concepts, selecting winners based on a pairwise dominance relation are an important structure often used in sports, as well as elections, and argumentation theory. Manipulation of such choice rules by coalitions of agents are a significant problem in most common rules. We deal with the problem of the manipulation of randomized choice rules by coalitions varying from a single agent, to two or more agents. We define two notions of coalitional manipulations of such choice rules based on whether or not utility is transferable. We show useful choice rules satisfying both notions of non-manipulability, and for the transferable utility case provide bounds on the level of Condorcet consistency.

We study methods to specify preferences among subsets of a set (a universe ). The methods we focus on are of two types. The first one assumes the universe comes with a preference relation on its elements and attempts to lift that relation to subsets of the universe. That approach has limited expressivity but results in orderings that capture interesting general preference principles. The second method consists of developing formalisms allowing the user to specify "atomic" improvements, and generating from them preferences on the powerset of the universe. We show that the particular formalism we propose is expressive enough to capture the lifted preference relations of the first approach, and generalizes propositional CP-nets. We discuss the importance of domain-independent methods for specifying preferences on sets for knowledge representation formalisms, selecting the formalism of argumentation frameworks as an illustrative example.

In this paper, we address the problem of change in an abstract argumentation system. We focus on a particular change: the addition of a new argument which interacts with previous arguments. We study the impact of such an addition on the outcome of the argumentation system, more particularly on the set of its extensions. Several properties for this change operation are defined by comparing the new set of extensions to the initial one, these properties are called structural when the comparisons are based on set-cardinality or set-inclusion relations. Several other properties are proposed where comparisons are based on the status of some particular arguments: the accepted arguments; these properties refer to the evolution of this status during the change, e.g., Monotony and Priority to Recency. All these properties may be more or less desirable according to specific applications. They are studied under two particular semantics: the grounded and preferred semantics.

Since argumentation is an inherently dynamic process, it is of great importance to understand the effect of incorporating new information into given argumentation frameworks. In this work, we address this issue by analyzing equivalence between argumentation frameworks under the assumption that the frameworks in question are incomplete, i.e. further information might be added later to both frameworks simultaneously. In other words, instead of the standard notion of equivalence (which holds between two frameworks, if they possess the same extensions), we require here that frameworks F and G are also equivalent when conjoined with any further framework H. Due to the nonmonotonicity of argumentation semantics, this concept is different to (but obviously implies) the standard notion of equivalence. We thus call our new notion strong equivalence and study how strong equivalence can be decided with respect to the most important semantics for abstract argumentation frameworks. We also consider variants of strong equivalence in which we define equivalence with respect to the sets of arguments credulously (or skeptically) accepted, and restrict strong equivalence to augmentations H where no new arguments are raised.

Abstract argumentation frameworks have received a lot of interest in recent years. Most computational problems in this area are intractable but several tractable fragments have been identified. In particular, Dunne showed that many problems can be solved in linear time for argumentation frameworks of bounded tree-width. However, these tractability results, which were obtained via Courcelle’s Theorem, do not directly lead to efficient algorithms. The goal of this paper is to turn the theoretical tractability results into efficient algorithms and to explore the potential of directed notions of tree-width for defining larger tractable fragments.

Argumentation is now a very fertile area of research in Artificial Intelligence. Yet, most approaches to reasoning with arguments in AI are based on a normative perspective, relying on intuition as to what constitutes correct reasoning, sometimes aided by purpose-built hypothetical examples. For these models to be useful in agent-human argumentation, they can benefit from an alternative, positivist perspective that takes into account the empirical reality of human reasoning. To give a flavour of the kinds of lessons that this methodology can provide, we report on a psychological study exploring simple reinstatement in argumentation semantics. Empirical results show that while reinstatement is cognitively plausible in principle, it does not yield full recovery of the argument status, a notion not captured in Dung's classical model. This result suggests some possible avenues for research relevant to making formal models of argument more useful.

In this work we will present an integration of a query-answering argumentation approach with an abstract agent programming language. Agents will argumentatively reason via queries, using information of their mental components. Special context-based queries will be used to model the interaction between mental components. Deliberation and execution semantics of the proposed integration are presented.

An understanding of the policy and resource availability constraints under which others operate is important for effectively developing and resourcing plans in a multi-agent context. Such constraints (or norms) are not necessarily public knowledge, even within a team of collaborating agents. What is required are mechanisms to enable agents to keep track of who might have and be willing to provide the resources required for enacting a plan by modeling the policies of others regarding resource use, information provision, etc. We propose a technique that combines machine learning and argumentation for identifying and modeling the policies of others. Furthermore, we demonstrate the utility of this novel combination of techniques through empirical evaluation.