Lecture Notes in Artificial Intelligence 4946 EditedbyJ.G.CarbonellandJ.Siekmann Subseries of Lecture Notes in Computer Science IyadRahwan SimonParsons ChrisReed(Eds.) Argumentation in Multi-Agent Systems 4th International Workshop, ArgMAS 2007 Honolulu, HI, USA, May 15, 2007 Revised Selected and Invited Papers 1 3 SeriesEditors JaimeG.Carbonell,CarnegieMellonUniversity,Pittsburgh,PA,USA JörgSiekmann,UniversityofSaarland,Saarbrücken,Germany VolumeEditors IyadRahwan BritishUniversityinDubai,FacultyofInformatics P.O.Box502216,Dubai,UAE E-mail:[email protected] SimonParsons CityUniversityofNewYork,BrooklynCollege DepartmentofComputerandInformationScience 2900BedfordAvenue,Brooklyn,NY11210,USA E-mail:[email protected] ChrisReed UniversityofDundee,DepartmentofAppliedComputing DundeeDD14HN,UK E-mail:[email protected] LibraryofCongressControlNumber:2008923358 CRSubjectClassification(1998):I.2.11,I.2,C.2.4,H.5.2-3 LNCSSublibrary:SL7–ArtificialIntelligence ISSN 0302-9743 ISBN-10 3-540-78914-6SpringerBerlinHeidelbergNewYork ISBN-13 978-3-540-78914-7SpringerBerlinHeidelbergNewYork Thisworkissubjecttocopyright.Allrightsarereserved,whetherthewholeorpartofthematerialis concerned,specificallytherightsoftranslation,reprinting,re-useofillustrations,recitation,broadcasting, reproductiononmicrofilmsorinanyotherway,andstorageindatabanks.Duplicationofthispublication orpartsthereofispermittedonlyundertheprovisionsoftheGermanCopyrightLawofSeptember9,1965, initscurrentversion,andpermissionforusemustalwaysbeobtainedfromSpringer.Violationsareliable toprosecutionundertheGermanCopyrightLaw. SpringerisapartofSpringerScience+BusinessMedia springer.com ©Springer-VerlagBerlinHeidelberg2008 PrintedinGermany Typesetting:Camera-readybyauthor,dataconversionbyScientificPublishingServices,Chennai,India Printedonacid-freepaper SPIN:12249947 06/3180 543210 Preface This volume presents the latest developments in the growing area of research at the interface of argumentationtheory and multiagent systems. This area has grown tremendously with many papers appearing in the recent special issue of the Artificial Intelligence Journal on “Argumentation” and the special issue of IEEE Intelligent Systems on “Argumentation Technologies.” Over the last few years, argumentation has been gaining increasing impor- tance in multiagent systems, mainly as a vehicle for facilitating rational inter- action(i.e.,interactionwhich involvesthe givingandreceivingofreasons).This is because argumentation provides tools for designing, implementing, and ana- lyzing sophisticated forms of interaction among rational agents. Argumentation hasmadesolidcontributionstothepracticeofmultiagentdialogues.Application domains include: legal disputes, business negotiation, labor disputes, team for- mation, scientific inquiry, deliberative democracy, ontology reconciliation, risk analysis, scheduling, and logistics. A single agent may also use argumentation techniquestoperformitsindividualreasoningbecauseitneedstomakedecisions under complex preferences policies, in a highly dynamic environment. Most papers in this volume appeared in the proceedings of the 4th Inter- national Workshop on Argumentation in Multiagent Systems (ArgMAS 2007), which took place in Honolulu, USA, in conjunction with the InternationalJoint Conference on Autonomous Agents and Multiagent Systems (AAMAS). This continues the success of the ArgMAS workshop series, which took place in tan- demwithAAMASinNewYorkin2004,Utrechtin2005,andHakodatein2006. Often we have invited papers on the topic of argumentation in multiagent systems from the main AAMAS conference as well as other major conferences for the givenyear,inorderto bringtogether the verybest ofthe year’sworkon argumentationinMASintoasinglevolume.Thistime,weinvitedrevisedpapers on argumentation in MAS from both AAMAS 2007 and AAAI 2007. These additional contributions were selected on the basis of their scientific quality and relevance to the topics emphasized here. Our objective has been to offer a comprehensiveandup-to-dateoverviewofthisrapidlyevolvinglandscape,aswe did in the previous volumes of this series which were all published by Springer (LNAI 3366, LNAI 4049, and LNAI 4766). Thebookisorganizedintothreeparts,eachaddressinganimportantproblem inargumentationandmultiagentsystems.PartI focuses onissuespertaining to dialogue. This part opens with an invited paper from AAMAS 2007 by Leila Amgoud,YannisDimopoulos,andPavlosMoraitis.Thispaperattemptsto gen- eralize and unify various models of argument-based negotiation (ABN) found in the literature. The second paper, by Iyad Rahwan, Philippe Pasquier, Liz Sonenberg, and Frank Dignum, formally explores some intuitions behind the usefulness of ABN models. Most work on argumentation addresses two-party dialogues.To address dialogues involving more than two agents, in an agent so- VI Preface ciety, Enrico Oliva, Peter McBurney, and Andrea Omicini present a framework based on the notion of co-argumentationartifact. The fourth paper in Part I is an invited paper by Simon Parsons, Peter McBurney,Elizabeth Sklar,and MichaelWooldridge,whichaddressesthe ques- tion of relevance in dialogues. The authors show how relevance can help agents decide which arguments to present (i.e., how to carry out a dialogue according to a given protocol). The following paper, by Laurent Perrussel, Sylvie Doutre, Jean-MarcTh´evenin,andPeterMcBurney,presentsaprotocolthatallowsagents to argue about the right to access information in a given context. Recently, an argument interchange format (AIF) has been proposed by the “argumentation in AI” community. However, while the AIF is quite mature in its representationofstaticargumentstructures,its abilityto captureargument- baseddialoguesisstillunderdeveloped.PartIconcludeswithapaperbySanjay Modgil and Jarred McGinnis that addresses this issue. PartIIfocusesonusingargumentationtoautomateorsupportvarioussingle- agent reasoning tasks. This part opens with an invited AAAI 2007 paper by Nicola´s D. Rotstein, Alejandro J. Garc´ıa, and Guillermo R. Simari. This paper describeshowargumentationcanbeusedtomechanizereasoninginbeliefdesire intention (BDI) agents.In a multiagent system, such BDI agents are faced with possiblyconflictingsocialnorms.Toaddressthe conflictbetweennorms,beliefs, desires, and intentions, the following paper, by Dorian Gaertner and Francesca Toni, presents a framework grounded in assumption-based argumentation. In the third paper in this part, Maxime Morge presents a framework for practical reasoning based on argumentation. Part II then closes with a paper, by Ca´ssia Trojahn, Paulo Quaresma, Renata Vieira, on using argumentationas a model for mapping multiple conflicting ontologies. Part III addresses an exciting new area in argumentation research, namely, the relationship between models of argumentation and models of learning. The section opens with an invited paper from AAMAS 2007, by Leila Amgoud and Mathieu Serrurier, on using argumentation to perform classification tasks nor- mally found in the machine learning literature. This is followedby another AA- MAS2007invitedpaper,bySantiOntan˜o´nandEnricPlaza,whichexploresthe use of argumentation to facilitate dialogue among cased-based reasoning agents who want to learn jointly from past cases. This part is wrapped up by two pa- pers by Wataru Makiguchi and Hajime Sawamura on the relationship between symbolic and neural network-based models of argumentation. We conclude this preface by extending our gratitude to the members of the Steering Committee and members of the Program Committee, who together helped make the ArgMAS workshop a success. We also thank the authors for theirenthusiasmtosubmitpaperstotheworkshop,andforrevisingtheirpapers on time for inclusion in this book. December 2007 Iyad Rahwan Simon Parsons Chris Reed Organization Program Chairs Simon Parsons City University of New York, USA Iyad Rahwan British University in Dubai, UAE (Fellow) University of Edinburgh, UK Chris Reed University of Dundee, UK ArgMAS Steering Committee Antonis Kakas University of Cyprus, Cyprus Nicolas Maudet Universit´e Paris Dauphine, France Peter McBurney University of Liverpool, UK Pavlos Moraitis Paris Descartes University, France Simon Parsons City University of New York, USA Iyad Rahwan British University in Dubai, UAE (Fellow) University of Edinburgh, UK Chris Reed University of Dundee, UK Program Committee Leila Amgoud IRIT, France Katie Atkinson University of Liverpool, UK Jamal Bentahar Laval University, Canada Guido Boella Universita` di Torino, Italy Brahim Chaib-draa Laval University, Canada Carlos Chesnevar Universitat de Lleida, Spain Frank Dignum Utrecht University, The Netherlands Rogier van Eijk Utrecht University, The Netherlands Frank Guerin University of Aberdeen, UK Joris Hulstijn Utrecht University, The Netherlands Anthony Hunter University College, London, UK Antonis Kakas University of Cyprus, Cyprus Nikos Karacapilidis University of Patras, Greece Nishan Karunatillake University of Southampton, UK Nicolas Maudet Universit´e Paris Dauphine, France Peter McBurney University of Liverpool, UK Jarred McGinnis Royal Holloway, University of London, UK Sanjay Modgil Cancer Research UK Pavlos Moraitis Paris Descartes University, France VIII Organization Bernard Moulin Laval University, Canada Søren Holbech Nielsen Aalborg University, Denmark Tim Norman University of Aberdeen, UK Nir Oren University of Aberdeen, UK Fabio Paglieri ISTC-CNR, Rome, Italy Xavier Parent King’s College, UK Simon Parsons City University of New York, USA Philippe Pasquier University of Melbourne, Australia Enric Plaza Spanish Scientific Research Council, Spain Henri Prade IRIT, Toulouse, France Henry Prakken Utrecht University, The Netherlands Alun Preece University of Aberdeen, UK Iyad Rahwan British University in Dubai, UAE (Fellow) University of Edinburgh, UK Sarvapali Ramchurn University of Southampton, UK Chris Reed University of Dundee, UK Michael Rovatsos University of Edinburgh, UK Hajime Sawamura Niigata University, Japan Sandip Sen University of Tulsa, USA Guillermo Simari Universidad Nacional del Sur, Argentina Elizabeth Sklar City University of New York, USA Katia Sycara Carnegie Mellon University, USA Francesca Toni Imperial College, London, UK Leon van der Torre University of Luxembourg, Luxembourg Paolo Torroni Universita` di Bologna, Italy Bart Verheij Maastricht University, The Netherlands Gerard Vreeswijk Utrecht University, The Netherlands Doug Walton University of Winnipeg, Canada Simon Wells University of Dundee, UK Steven Willmott Universitat Polit´ecnica de Catalunya, Spain Mike Wooldridge University of Liverpool, UK Table of Contents Part I: Argumentation and Dialogue A General Framework for Argumentation-Based Negotiation........... 1 Leila Amgoud, Yannis Dimopoulos, and Pavlos Moraitis OntheBenefitsofExploitingHierarchicalGoalsinBilateralAutomated Negotiation ..................................................... 18 Iyad Rahwan, Philippe Pasquier, Liz Sonenberg, and Frank Dignum Co-argumentationArtifact for Agent Societies ....................... 31 Enrico Oliva, Peter McBurney, and Andrea Omicini On the Relevance of Utterances in Formal Inter-agent Dialogues ....... 47 Simon Parsons, Peter McBurney, Elizabeth Sklar, and Michael Wooldridge A PersuasionDialog for Gaining Access to Information ............... 63 Laurent Perrussel, Sylvie Doutre, Jean-Marc Th´evenin, and Peter McBurney Towards Characterising Argumentation Based Dialogue in the Argument Interchange Format..................................... 80 Sanjay Modgil and Jarred McGinnis Part II: Argument-Based Reasoning Preferences and Assumption-Based Argumentation for Conflict-Free Normative Agents................................................ 94 Dorian Gaertner and Francesca Toni TheHedgehogandtheFox:AnArgumentation-BasedDecisionSupport System ......................................................... 114 Maxime Morge An Extended Value-Based Argumentation Framework for Ontology Mapping with Confidence Degrees.................................. 132 Ca´ssia Trojahn, Paulo Quaresma, and Renata Vieira Defeasible Argumentation Support for an Extended BDI Architecture..................................................... 145 Nicola´s D. Rotstein, Alejandro J. Garc´ıa, and Guillermo R. Simari X Table of Contents Part III: Argumentation and Learning Arguing and Explaining Classifications ............................. 164 Leila Amgoud and Mathieu Serrurier An Argumentation-Based Framework for Deliberation in Multi-agent Systems ........................................................ 178 Santi Ontan˜´on and Enric Plaza A Hybrid Argumentation of Symbolic and Neural Net Argumentation (Part I)......................................................... 197 Wataru Makiguchi and Hajime Sawamura A Hybrid Argumentation of Symbolic and Neural Net Argumentation (Part II) ........................................................ 216 Wataru Makiguchi and Hajime Sawamura Author Index.................................................. 235 A General Framework for Argumentation-Based Negotiation Leila Amgoud1, Yannis Dimopoulos2, and Pavlos Moraitis3 1 IRIT– CNRS,118, route de Narbonne,31062 Toulouse, France [email protected] 2 University of Cyprus, 75 Kallipoleos Str. PO Box 20537, Cyprus [email protected] 3 Paris-Descartes University,45 rue des Saints-Peres, 75270 Paris, France [email protected] Abstract. This paper proposes a unified and general framework for argumentation-based negotiation, in which the role of argumentation is formally analyzed. The framework makes it possible to study the out- comes of an argumentation-based negotiation. It shows what an agree- ment is,howitisrelatedtothetheoriesoftheagents,whenitispossible, and how this can be attained by the negotiating agents in this case. It defines also the notion of concession, and shows in which situation an agent will make one, as well as how it influences the evolution of the dialogue. Keywords: Argumentation, Negotiation. 1 Introduction Roughly speaking, negotiation is a process aiming at finding some compromise or consensus between two or several agents about some matters of collective agreement,suchaspricingproducts,allocatingresources,orchoosingcandidates. Negotiationmodelshavebeenproposedforthedesignofsystemsabletobargain in an optimal way with other agents for example, buying or selling products in e-commerce. Different approaches to automated negotiation have been investigated, in- cluding game-theoretic approaches (which usually assume complete informa- tion and unlimited computation capabilities) [11], heuristic-based approaches whichtrytocopewiththeselimitations[6],andargumentation-based approaches [2,3,7,8,9,12,13]whichemphasizethe importanceofexchanginginformationand explanations between negotiating agents in order to mutually influence their behaviors (e.g. an agent may concede a goal having a small priority), and con- sequently the outcome of the dialogue.Indeed, the two first types of settings do notallowfortheadditionofinformationorforexchangingopinionsaboutoffers. Integratingargumentationtheoryinnegotiationprovidesa goodmeansforsup- plying additional information and also helps agents to convince each other by adequate arguments during a negotiation dialogue. Indeed, an offer supported I.Rahwan,S.Parsons,andC.Reed(Eds.):ArgMAS2007,LNAI4946,pp.1–17,2008. (cid:2)c Springer-VerlagBerlinHeidelberg2008