[Date Prev][Date Next][Date Index]

December 7th/8th: Workshop on Argumentation

Dear Colleagues and Friends,

On 7th and 8th December, we will have a small international meeting
on abstract argumentation at our institute with several interesting
talks. The full programme with abstracts is attached.

The meeting will take place in the
Seminarroom "von Neumann" (entrance via the yard),
Favoritenstrasse 9-11, 1040 Wien.

Feel free to join the meeting at any time
feasible for you!

Best wishes,
Stefan Woltran.
Argumentation Christmas Meeting, December 7-8, 2010
Vienna University of Technology

Seminarroom "von Neumann" 
(entrance via the yard)
Favoritenstrasse 9-11, 1040 Wien

Tue, 7.12.
9:00-9:30 G. Brewka:      Abstract Coalition Frameworks: What Coalition Formation Can 
	  	          Learn from Argumentation (Work in Progress)
9:30-10:30 S. Villata:    Argumentation Patterns: Examples and Challenges
10:30-11:00 Coffee
11:00-11:45 W. Dvorak:    On the Intertranslatability of Argumentation Semantics
11:45-14:00 Lunch
14:00-14:30 S. Szeider:   Computational Aspects of Argumentation: Parameters and 
                          Kernels (Work in Progress)
14:30-15:00 R. Baumann:   Splitting of Argumentation Frameworks
15:00-15:30 Coffee
15:30-16:00 S. Gaggl:     Strong Equivalence for CF semantics (Work in Progress) 
16:00-16:30 E. Oikarinen: Local Equivalence for Grounded Semantics

Wed, 8.12.
9:00-10:30  M. Caminada: Putting Things Together; Open Research Challenges in 
                         Argumentation Theory
10:30-11:00 coffee
11:00-12:15 S. Modgil:   Instantiations of Extended Argumentation Frameworks and their 
                         Satisfaction of Rationality Postulates
12:15-14:15 lunch
14:15-15:15 S. Woltran:  The WWTF Project "New Methods for Analyzing, Comparing, and Solving 
                         Argumentation Problems"; Progress Report and Lessons Learned.
15:15-      Discussion:  Future Directions of Argumentation.

List of Abstracts (Incomplete)

Gerd Brewka: Abstract Coalition Frameworks: What Coalition Formation Can
             Learn from Argumentation (Work in Progress) 

In the talk we discuss to which extent an abstract theory of coalition formation can benefit
from abstract argumentation. Starting from graphs which are similar to argumentation
frameworks, we interpret the set of nodes as agents and the binary relation among nodes as
threat. Notions like conflict-freeness, admissibility, satility have natural counterparts
under this new reading. However, whereas conflict-freeness still seems to be an important
notion, the role of admissibility is less clear, as we illustrate using several examples.
Rather than singling out a particular property desired coalitions should satisfy, we
introduce an abstract strength relation on coalitions. Based on this strength relation we
define a notion of equilibrium, somewhat reminiscent of Nash equilibria in strategic games,
for partitions of agents into coalitions.

Serena Villata: Argumentation Patterns; Examples and Challenges

Argumentation patterns are general reusable solutions to commonly occurring problems in the
design of argumentation frameworks, such as the support of a claim by data
in the Toulmin scheme. I introduce a formal framework for the semantics of argumentation
patterns describing relationships and interactions among arguments, without
instantiating the involved abstract arguments. Moreover, I show how to define argumentation
patterns using a two sorted argumentation framework where focal arguments are distinguished
from auxiliary arguments, and I show how to compute their semantics by reusing Dung's
standard semantics. Whereas ultimately arguments must be instantiated, in my experience of
modeling argumentation, there is a need at the abstract level to define argumentation patterns.
A case study is proposed to illustrate how argumentation patterns can be
used to analyze and define an argumentation framework in a modular way, and how they support
reuse. The challenge is to apply argumentation patterns to normative problems, to the
representation of trust and, finally, to decision making. Examples explaining these 
challenges will be provided. 

Wolfgang Dvorak: On the Intertranslatability of Argumentation Semantics

Translations between different nonmonotonic formalisms always have
been an important topic, in particular to understand the
knowledge representation capabilities those formalisms offer.
We provide such an investigation in terms of different semantics
proposed for abstract argumentation frameworks. Although the properties
of these different semantics are nowadays well understood, there are no 
explicit results about intertranslatability.  We provide such translations 
wrt. different properties and also give some negative results that show 
that certain translations are not possible.

Stefan Szeider: Computational Aspects of Argumentation; Parameters and Kernels (Work in Progress)

Parameterized Complexity (PC) is a new foundational framework for computational complexity
and algorithms that takes structural aspects of problem instances into account. It allows a
more fine-grained analysis than the classical P vs NP setting. In this talk I will outline
the basic concepts of PC and highlight the adequacy of PC for the study of computational
problems that arise in the area of abstract argumentation. I will review some recent results,
including a new kernelization-based algorithm for finding large extensions in argumentation

Ringo Baumann: Splitting of Argumentation Frameworks

Splitting Results in non-mononotonic formalisms have a long tradition. On the one hand, 
these results can be used to improve existing computational procedures, and on the other 
hand they yield deeper theoretical insights into how a non-monotonic approach works. 
In the 90's Lifschitz and Turner proved splitting results for logic programs and
default theory. In this paper we establish similar results for Dung style argumentation 
frameworks (AFs) under the most important semantics, namely stable, preferred, complete 
and grounded semantics. Furthermore we show how to use these results in dynamical 

Sarah Gaggl: Strong Equivalence for CF semantics (Work in Progress) 

In this talk we present our ongoing work about strong equivalence with
respect to conflict-free semantics. Till now, strong equivalence for
argumentation frameworks has been analyzed for several semantics which
are all admissible based. Semantics which are not based on the notion of
defence have been neglected so far. To close this gap, we will show some
recently achieved results with respect to maximal conflict-free sets,
stage  and cf2 extensions.

Emilia Oikarinen: Local Equivalence for Grounded Semantics

Argumentation is an inherently dynamic process where new arguments
are raised or new relations between arguments are recognized. Thus it is of
great importance to understand the effect of incorporating new information
into given argumentation frameworks. In a recent paper this issue was
addressed 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. Two
notions of equivalence were introduced: strong equivalence requires here
that frameworks possess the same extensions when conjoined with any further
framework, while local equivalence restricts strong equivalence to
augmentations with frameworks where no new arguments are raised. In this
talk, we focus in detail the specifics of characterization of the local
equivalence under the grounded semantics.

Martin Caminada: Putting Things Together; Open Research Challenges in Argumentation Theory

Although much research has been done on graph-theoretical notions of argumentation, much less
has been done on how the resulting abstract theory can be instantiated for the purpose of
logical entailment. This is surprising, because many argumentation researchers would rather
describe themselves as "logicians" instead "graph theoreticians". The question of how
entailment fits into abstract argumentation theory is, however, a difficult one, as only very
little of the graph-theoretical argumentation research has been done with entailment in mind.
In many cases it is simply not clear how the abstract work can be used for any reasonable
form of logical entailment at all. In this presentation we point out what the issues are and
make some suggestions for putting things right.

Sanjay Modgil: Instantiations of Extended Argumentation Frameworks and their Satisfaction of 
               Rationality Postulates

Recently, Prakken extended the ASPIC framework so as to define structured Dung frameworks
that adopt a level of abstraction intermediate between Dung's original framework and concrete
instantiating logics. One could thus meaningfully study and show satisfaction of Caminada and
Amgoud's rationality postulates for a broad range of concrete logics instantiating structured
Dung frameworks.

In a separate development, I defined Extended Argumentation Frameworks (EAFs) that augment
Dung theory so that arguments can attack attacks as well as arguments. EAF thus provided an
abstract dialectical semantics for instantiating logics that accommodate reasoning about
preferences/priorities . Although I have studied instantiations of EAFs , their level of
abstraction, as with the original Dung frameworks, precludes a systematic analysis of
concrete instantiations and their satisfaction of the aforementioned rationality postulates.

To address this, Prakken and I have recently further extended the ASPIC framework so as to
define structured EAFs. We show how, and under what conditions, the rationality postulates
are satisfied by a broad range of logics that instantiate structured EAFs. Our work thus
provides for principled development of instantiating logics that accommodate object level
reasoning about priorities, including classical logic based approaches to argumentation,
assumption based argumentation, and logic programming with defeasible priorities.

In my talk I will present the above work and discuss some of its implications.

Stefan Woltran: The WWTF Project "New Methods for Analyzing, Comparing, and Solving 
                Argumentation Problems"; Progress Report and Lessons Learned.

We give an overview about the work plan of the 3-years project "New Methods for Analyzing, 
Comparing, and Solving Argumentation Problems", funded by the Wiener Wissenschafts-, Forschungs- 
und Technologiefonds (WWTF) under grant ICT 08-028. The main research questions within 
the project are (i) Theoretical Underpinnings for Comparison and Simplification; 
(ii) Theoretical Tractability and its Realization; (iii) Implementation; and (iv) 
Integration and Evaluation. Work packages (i)--(iii) have been already addressed in 
a satisfactorily way, and we shall highlight some of the achieved results in this talk.
Moreover, it is still an open issue how the evaluation of argumentation systems should be
done in particular under the absence of benchmark collections.