Jean-Guy Mailly

Recent advancements in algorithms for abstract argumentation make it possible now to solve reasoning problems even with argumentation frameworks of large size, as demonstrated by the results of the various editions of the International Competition on Computational Models of Argumentation (ICCMA). However, the solvers participating to the competition may be hard to use for non-expert programmers, especially if they need to incorporate these algorithms in their own code instead of simply using the command-line interface. Moreover, some ICCMA solvers focus on the ICCMA tracks, and do not implement algorithms for other problems. In this paper we describe pygarg, a Python implementation of the SAT-based approach used in the argumentation solver CoQuiAAS. Contrary to CoQuiAAS and most of the participants to the various editions of ICCMA, pygarg incorporates all problems that have been considered in the main track of any edition of ICCMA. We show how to easily use pygarg via a command-line interface inspired by ICCMA competitions, and then how it can be used in other Python scripts as a third-party library.

The International Competition on Computational Models of Argumentation (ICCMA) focuses on reasoning tasks in abstract argumentation frameworks. Submitted solvers are tested on a selected collection of benchmark instances, including artificially generated argumentation frameworks and some frameworks formalizing real-world problems. This paper presents the novelties introduced in the organization of the Third (2019) and Fourth (2021) editions of the competition. In particular, we proposed new tracks to competitors, one dedicated to dynamic solvers (i.e., solvers that incrementally compute solutions of frameworks obtained by incrementally modifying original ones) in ICCMA’19 and one dedicated to approximate algorithms in ICCMA’21. From the analysis of the results, we noticed that i) dynamic recomputation of solutions leads to significant performance improvements, ii) approximation provides much faster results with satisfactory accuracy, and iii) classical solvers improved with respect to previous editions, thus revealing advancement in state of the art.

In a recently published book, the French writer and comedian François Rollin has discussed various aspects of the notion of stupidity, including artificial stupidity, the stupid counterpart of artificial intelligence. His claim is that a system of artificial stupidity is a system that provides wrong answers to any task it should solve, leading to absurd solutions in most cases. We believe that this claim is (at least partially) false and that designing artificial stupidity is not as trivial as it seems. In this article, we discuss why and how one could design a system of artificial stupidity. We believe that such a reflection on (artificial) stupidity can bring about some interesting insights about (artificial) intelligence.

Operations like belief change or merging have been adapted to the context of abstract argumentation. However, these operations may require to express some uncertainty or some disjunction in the result, which is not representable in classical AFs. For this reason, some of these earlier works require a set of AFs or a set of extensions as the outcome of the operation, somehow to represent a “disjunction” of AFs or extensions. In parallel, the notion of Incomplete AFs (IAFs) has been developed recently. It corresponds to AFs where the existence of some arguments or attacks may be uncertain. Each IAF can be associated with a set of classical AFs called completions, that correspond to different ways of resolving the uncertainty. While these IAFs could be good candidates for a compact representation of a disjunction of AFs, we prove that this model is not expressive enough. Then we introduce Constrained IAFs, that include a propositional formula allowing to select the set of completions used for reasoning. We prove that this model is expressive enough for representing any set of AFs, or any set of extensions. Moreover, we study the complexity of various decision problems related to the verification of extensions and the acceptability of arguments. While some of them are one level higher in the polynomial hierarchy (compared to their counterpart with standard IAFs), most of them have the same complexity than in the case of IAFs. Finally, we show that CIAFs can be used to model a new form of extension enforcement, where the possible evolutions of an AF are taken into account and modeled by the completions of the CIAF.

In the realm of multi-agent systems, dialogues rooted in argumentation (such as those for persuasion and negotiation) rely on the exchange of arguments among autonomous agents. These agents must continually re-evaluate the acceptability of arguments during each step of the dialogue. The challenge lies in finding efficient computational methods due to the inherent complexity of traditional reasoning tasks in argumentation. In this study, we propose an innovative approach that leverages modern SAT solving techniques to dynamically reassess the acceptability status credulous or skeptical) of arguments, adhering to various classical semantics found in existing literature. Our method capitalizes on the assumption mechanism prevalent in contemporary SAT solvers. While previous approaches to argumentation dynamics have also utilized this mechanism, our method stands out by making minimal assumptions, rendering it highly practical. Through comprehensive experimental evaluation, our approach demonstrates robust scalability and superior performance when compared to existing methods.

The growing interest in generalizations of Dung’s abstract argumentation frameworks has recently led to the simultaneous and independent discovery of a combination of two of these generalizations: Bipolar Argumentation Frameworks (BAFs), where a relation representing supports between arguments is added, and Incomplete Argumentation Frameworks (IAFs), where the existence of arguments and attacks may be uncertain, resulting in the so-called Incomplete Bipolar Abstract Argumentation Frameworks (IBAFs). This paper digs deeper into such a combination by: (i) providing a thoughtful analysis of the existing notions of completion (the hypothetical removal of uncertainty used in IBAFs to reason about argument acceptability); (ii) proposing, motivating and studying new notions of completion; (iii) throwing new complexity results on argument acceptability problems associated with IBAFs; (iv) encoding these reasoning problems into a lightweight version of dynamic logic.