[Back]

W. Dvorak, M. König, S. Woltran:
"Graph-Classes of Argumentation Frameworks with Collective Attacks";
Talk: 17th European Conference on Logics in Artificial Intelligence (JELIA 2021), Klagenfurt; 2021-05-21 - 2021-05-25; in: "Logics in Artificial Intelligence - 17th European Conference, {JELIA} 2021, Virtual Event, May 17-20, 2021, Proceedings", (2021), 3 - 17.

Argumentation frameworks with collective attacks (SETAFs)
have gained increasing attention in recent years as they provide a natural extension of the well-known abstract argumentation frameworks (AFs)due to Dung. Concerning complexity, it is known that for the standard reasoning tasks in abstract argumentation, SETAFs show the same behavior as AFs, i.e. they are mainly located on the rst or second level of the polynomial hierarchy. However, while for AFs there is a rich literature on easier fragments, complexity analyses in this direction are still missing for SETAFs. In particular, the well-known graph-classes of acyclic AFs, even-cycle-free AFs, symmetric AFs, and bipartite AFs have been shown tractable. In this paper, we aim to extend these results to the more general notion of SETAFs. In particular, we provide various syntactic notions on SETAFs that naturally generalize the graph properties for directed hypergraphs, and perform a complexity analysis of the prominent credulous and skeptical acceptance problems for several different widely used semantics.

Abstract argumentation; complexity; SETAF; collective attacks

http://dx.doi.org/10.1007/978-3-030-75775-5_1

Project Head Stefan Woltran:
HYPAR

Project Head Stefan Woltran:
START