Publications Leon van der Torre

Five Guidelines for Normative Multiagent Systems

A Normative Multiagent Approach to Requirements Engineering

A Multimodal view on Access Control and Trust Management: Fibred Security Language

A complete conclusion-based procedure for judgment aggregation

Analyzing Cooperation in Iterative Social Network Design

Fibred Security Language

We study access control policies based on the says operator by introducing a logical framework called Fibred Security Language (FSL) which is able to deal with features like joint responsibility between sets of principals and to identify them by means of first-order formulas. FSL is based on a multimodal logic methodology. We first discuss the main contributions from the expressiveness point of view, we give semantics for the language both for classical and intuitionistic fragment), we then prove that in order to express well-known properties like Ôspeaks-forÕ or Ôhand-offÕ, defined in terms of says, we do not need second-order logic (unlike previous approaches) but a decidable fragment of first-order logic suffices. We propose a model-driven study of the says axiomatization by constraining the Kripke models in order to respect desirable security properties, we study how existing access control logics can be translated into FSL and we give completeness for the logic.

springer

Representing Excuses in Social Dependence Networks

The Role of Goals in Belief Selection

Guido Boella, C«elia da Costa Pereira , Gabriella Pigozzi , Andrea Tettamanzi , and Leendert van der Torre

Algorithms for finding coalitions exploiting a new reciprocity condition

Guido Boella, Luigi Sauro and Leendert van der Torre

Abstract PDF

Dynamics in Argumentation with Single Extensions: Abstraction Principles and the Grounded Extension

A Middleware for modeling Organizations and Roles in Jade Matteo Baldoni, Guido Boella, Valerio Genovese, Roberto Grenna, Andrea Mugnaini, and Leendert van der Torre

Convivial Ambient Technologies: Requirements, Ontology, and Design

Abstract pdf

Four Ways to Change Coalitions: Agents, Dependencies, Norms and Internal Dynamics

The Design of Convivial Multiagent Systems

A Normative Multiagent Approach to Requirements Engineering

Normative Framework for Normative System Change

Social Network Semantics for Agent Communication

Dynamics in Argumentation with Single Extensions

The Design of Convivial Multiagent Systems

The Interplay between Relationships, Roles and Objects

Social Viewpoints for Arguing about Coalitions

Changing Institutional Goals and Beliefs of Autonomous Agents

Reasoning about Constitutive Norms, Counts-As Conditionals, Institutions, Deadlines and Violations

Time and Defeasibility in FIPA ACL Semantics

Norm negotiation in online multi-player games

http://www.springerlink.com/openurl.asp?genre=article&id=doi:10.1007/s10115-008-0162-2.

Making Others Believe What They Want

We study the interplay between argumentation and belief revision within the MAS framework. When an agent uses an argument to persuade another one, he must consider not only the proposition supported by the argument, but also the overall impact of the argument on the beliefs of the addressee. Different arguments lead to different belief revisions by the addressee. We propose an approach whereby the best argument is defined as the one which is both rational and the most appealing to the addressee.

http://dx.doi.org/10.1007/978-0-387-09695-7_21

Self Adaptive Coalitions in Multiagent Systems

Coalitions are usually deÞned with respect to a static framework of dependencies among agents. In this paper we propose a dynamic view of dependence networks to en- able dynamic coalitions which can self adapt to a situation by exploiting the possibility to trigger other agentsÕ goals.

saso08.pdf

Conditional norms and dyadic obligations

How to Program Organizations and Roles in the JADE Framework

The organization metaphor is often used in the design and imple- mentation of multiagent systems. However, few agent programming languages provide facilities to deÞne them. Several frameworks are proposed to coordinate MAS with organizations, but they are not programmable with general purpose languages. In this paper we extend the JADE framework with primitives to pro- gram in Java organizations structured in roles, and to enable agents to play roles in organizations. Roles facilitate the coordination of agents inside an organiza- tion and offer new abilities (powers) in the context of organizations to the agents which satisfy the requirements necessary to play the roles. To program organiza- tions and roles, we provide primitives which enable an agent to enact a new role in an organization to invoke powers.

mates08.pdf http://dx.doi.org/10.1007/978-3-540-87805-6_4

A Normative View on the Blocks World

From belief change to preference change

Conclusion-Based Procedure for Judgment Aggregation Satisfying Premise Independence

E-business Outsourcing: a New Approach combining Information Rights Technologies and Agent-Based Theory on Norms

Independence in judgment aggregation

Goal Change in the Context of Beliefs: Partial Implication, Logical Properties, and Complexity

Partial implication semantics in the context of a background theory has been introduced to formalize partial goal satisfaction in the context of beliefs. In this paper, we introduce strong partial implication prohibiting redundancies and weak partial implication allowing side effects, we study their semantic as well as complexity properties, and we apply the three notions of partial implication to goal change in the context of beliefs.

aamas08.pdf

Conviviality Masks in Role-Based Institutions Multi-Agent Teleconferencing in Virtual Worlds

Merging Roles in Coordination and in Agent Deliberation

Reasoning About Norms, Obligations, Time and Agents

J. Broersen and L. van der Torre.

A Common Ontology of Agent Communication Languages: Modeling Mental Attitudes and Social Commitments using Roles

Guido Boella, Rossanna Damiano, Joris Hulstijn and Leendert van der Torre

There are two main traditions in defining a semantics for agent communication languages, based either on mental attitudes or on social commitments. These traditions share speech acts as operators with preconditions and effects, and agents playing roles like speaker and hearer, but otherwise they rely on distinct ontologies. They refer not only to either belief and intention or various notions of social commitment, but also to distinct speech acts and distinct kinds of dialogue. In this paper, we propose a common ontology for both approaches based on public mental attitudes attributed to role instances. Public mental attitudes avoid the unverifiability problem of private mental states, while reusing the logics and implementations developed for FIPA compliant approaches. Moreover, a common ontology of communication primitives allows for the construction of agents which do not need separate reasoning modules to participate in dialogues with both mental attitudes and social commitments compliant agents. Moreover, a common ontology of communication primitives allows for the construction of agents participating in and combining the full range of dialogues covered by the individual approaches without having to redefine the existing protocols to cope with new dialog types. We illustrate how to extend the ontology to a semantics for agent communication and how to define mappings from existing semantics to the new one.

ao07b.pdf

Roles, an Interdisciplinary Perspective

Guido Boella, Leendert van der Torre and Harko Verhagen

The notion of role is ubiquitous in many fields of computer science, like programming languages, software engineering, coordination languages, databases, multiagent systems, knowledge representation, formal ontology, computational linguistics, security, and conceptual modelling, and also outside computer science, like in cognitive science, organizational science and linguistics.

In computer science, the discussion about roles started in the '70s with Bachman and Daya (1977), and, with a recurring trend, it comes back to the attention of the research community. Recently, roles have been used in many areas to handle interaction, for example, role based access control in security with the RBAC model (Sandhu et al. (1996)), collaboration roles in UML to describe the interaction among classes (Rumbaugh et al. (2004)), channels connecting components in coordination languages (Arbab (2003)), the separation of concerns to describe the interaction properties of objects in new contexts in programming languages, etc.With the rise of the internet, new communication possibilities and interactive computing created a new demand of research about roles, for example, in organizations in open multiagent systems, in role based programming languages, in using roles for the composition of web services, and in defining roles in standards for interoperability.

Notwithstanding this revival of the research about the notion of role, little agreement seems possible among the proposals in the different fields. This lack of agreement leads to the impossibility of transferring the results from one area to the other, and even inside a single area, a consequence which is quite unpleasant in a moment where the sharing of knowledge and standardization represent an added value in many fields. For example, in ontology, the lack of a common definition of role constitutes a problem for the interconnection of different knowledge bases: the result is that a widely used ontology language like OWL does not consider roles as a primitive. In multiagent systems, the openness of systems highlights the need of commonly accepted definitions, but again without a common notion of role it is not possible for a new agent to become part of an organization to interact with other agents; and in programming languages, software reuse can be improved only by a more developed theory of how objects interact with each other basing on the roles they play.

The likely reasons of these divergences are that many papers on the notion of role fail to have an interdisciplinary character, that much work proposes new definitions of roles to deal with particular practical problems, and that role seems an intuitive notion which can be grasped in its prototypical characters, but it is instead a deceptive one when details must be clarified. Few proposals, like Steimann (2000) or Masolo et al. (2004), have a more general attitude, and try to find a problem independent definition of role and to formalize it.

ao07.pdf

Adding Roles to Relationship Patterns

Matteo Baldoni, Guido Boella and Leendert van der Torre

In this paper we study how roles can be added to patterns modelling relationships in Object Oriented programming, and which new relationship patterns can be introduced using roles. Relationships can be introduced in programming languages either by reducing them to attributes of the objects which participate in the relationship, or by modelling the relationship itself as a class whose instances have the participants of the relationships among their attributes. However, even if roles have been recognized as an essential component of relationships, also in modelling languages like UML, they have not been introduced in Object Oriented programming when it is necessary to model relationships. Introducing roles allows to add attributes and behaviors to the participants in the relationship, rather than to the relationship itself, and to distinguish natural types as classes participating in the relationships from the roles the participants acquire in the relationships. In this paper we show how the role model proposed in powerJava can be used to endow relationships with roles, both in the relationship as attribute and in the relationship object pattern. Finally, since these patterns have different advantages and limitations, we propose a third pattern based on roles which benefits from the advantages of the two previous patterns when modelling relationships.

woa07.pdf

What You Should Believe: Obligations and Beliefs

Guido Boella, Celia da Costa Pereira, Andrea Tettamanzi, Gabriella Pigozzi, and Leendert van der Torre

This paper presents and discusses a novel approach to indeterministic belief revision. An indeterministic belief revision operator assumes that, when an agent is confronted with a new piece of information, it can revise its belief base in more than one way. We define a rational agent not only in terms of what it believes, as often assumed in belief revision, but also of what it ought or is obliged to do. Hence, we propose that the agent's goals play a role in the choice of (possibly) one of the several available revision options. Properties of the new belief revision mechanism are also investigated.

dkb07.pdf

Roles in Coordination and in Agent Deliberation: A Merger of Concepts

Matteo Baldoni, Guido Boella. Valerio Genovese and Leendert van der Torre

Integrating Architectural Models

Farhad Arbab, Frank de Boer, Marcello Bonsangue, Marc Lankhorst, Erik Proper, and Leendert van der Torre

On the Acceptability of Incompatible Arguments

Souhila Kaci, Leendert van der Torre and Emil Weydert

In this paper we study the acceptability of incompatible arguments within Dung's abstract argumentation framework. As an example we introduce an instance of Dung's framework where arguments are represented by propositional formulas and an argument attacks another one when the conjunction of their representations is inconsistent, which we characterize as a kind of symmetric attack. Since symmetric attack is known to have the drawback to collapse the various argumentation semantics, we consider also two variations. First, we consider propositional arguments distinguishing support and conclusion. Second, we introduce a preference ordering over the arguments and we define the attack relation in terms of a symmetric incompatibility relation and the preference relation. We show how to characterize preference-based argumentation using a kind of acyclic attack relation.

ecsqaru07.pdf

Preference-based Argumentation: Arguments Supporting Multiple Values

Souhila Kaci and Leendert van der Torre

In preference-based argumentation theory, an argument may be preferred to another one when, for example, it is more specific, its beliefs have a higher probability or certainty, or it promotes a higher value. In this paper we generalize Bench-Capon's value-based argumentation theory such that arguments can promote multiple values, and preferences among values or arguments can be specified in various ways. We assume in addition that there is default knowledge about the preferences over the arguments, and we use an algorithm to derive the most likely preference order. In particular, we show how to use non-monotonic preference reasoning to compute preferences among arguments, and subsequently the acceptable arguments, from preferences among values. We show also how the preference ordering can be used to optimize the algorithm to construct the grounded extension by proceeding from most to least preferred arguments.

ijar07.pdf

An Attacker Model for Normative Multi-Agent Systems

Guido Boella and Leendert van der Torre

Substantive and Procedural Norms in Normative Multiagent Systems

Guido Boella and Leendert van der Torre

Procedural norms are instrumental norms addressed to agents playing a role in the nor- mative system, for example to motivate these role playing agents to recognize violations or to apply sanctions. Procedural norms have first been discussed in law, where they address legal practitioners such as legislators, lawyers and policemen, but they are dis- cussed now too in normative multiagent systems to motivate software agents. Procedural norms aim to achieve the social order specified using regulative norms like obligations and permissions, and constitutive norms like counts-as obligations. In this paper we formalize procedural, regulative and constitutive norms using input/output logic enriched with an agent ontology and an abstraction hierarchy. We show how our formalization explains Castelfranchi's notion of mutual empowerment, stating that not only the agents playing a role in a normative system are empowered by the normative system, but the normative system itself is also empowered by the agents playing a role in it. In our terminology, the agents are not only institutionally empowered, but they are also delegated normative goals from the system. Together, institutional empowerment and normative goal delega- tion constitute a mechanism which we call delegation of power, where agents acting on behalf of the normative system become in charge of recognizing which institutional facts follow from brute facts.

jal08.pdf

Power in Norm Negotiation

Guido Boella and Leendert van der Torre

In social mechanism design, norm negotiation creates individual or contractual obligations fulfilling goals of the agents. The social delegation cycle distinguishes among social goal negotiation, obligation and sanction negotiation and norm acceptance. Power may affect norm negotiation in various ways, and we therefore introduce a new formalization of the social delegation cycle based on power and dependence, without referring to the rule structure of norms, actions, decision variables, tasks, and so on.

kesamsta07.pdf

Reasoning With Various Kinds of Preferences: Logic, Non-Monotonicity, and Algorithms

Souhila Kaci and Leendert van der Torre

As systems dealing with preferences become more sophisticated, it becomes essential to deal with various kinds of preference statements and their interaction. We introduce a non-monotonic logic distinguishing sixteen kinds of preferences, ranging from strict to loose and from careful to opportunistic, and two kinds of ways to deal with uncertainty, either optimistically or pessimistically. The classification of the various kinds of preferences is inspired by a hypothetical agent comparing the two alternatives of a preference statement. The optimistic and pessimistic way of dealing with uncertainty correspond on the one hand to considering either the best or the worst states in the comparison of the two alternatives of a preference statement, and on the other hand to the calculation of least or most specific "distinguished" preference orders from a set of preference statements. We show that each way to calculate distinguished preference orders is compatible with eight kinds of preferences, in the sense that it calculates a unique distinguished preference order for a set of such preference statements, and we provide efficient algorithms that calculate these unique distinguished preference orders. In general, optimistic kinds of preferences are compatible with optimism in calculating distinguished preference orders, and pessimistic kinds of preferences are compatible with pessimism in calculating distinguished preference orders. However, these two sets of eight kinds of preferences are not exclusive, such that some kinds of preferences can be used in both ways to calculate distinguished preference orders, and other kinds of preferences cannot be used in either of them. We also consider the merging of optimistically and pessimistically constructed distinguished preferences orders.

Springer

Interaction between Objects in powerJava

Matteo Baldoni, Guido Boella and Leendert van der Torre

In this paper we start from the consideration that high level interaction between entities like web services has very different properties with respect to the interaction between objects at the lower level of programming languages in the object oriented paradigm. In particular, web services, for security, usability and user adaptability reasons, offer different operations to different users by means of access control and keep track of the state of the interaction with each user by means of sessions. The current vision in object orientation, instead, considers attributes and operations of objects as being objective and independent from the interaction with another object, which is sessionless. To introduce these features in the interaction between objects directly in object oriented programming languages, we take inspiration from how access control is regulated by means of roles. Roles allow objects to offer different operations depending on the type of the role, of the type and identity of the player of the role, and to define session-aware interaction. We start from a definition of roles given in ontologies and knowledge representation and we discuss how this definition of roles can be introduced in Java, building our language powerJava.

jot07.pdf

FIPA Communicative Acts in Defeasible Logic

Guido Boella, Joris Hulstijn, Regis Riveret, Guido Governatori, Antonino Rotolo, and Leendert van der Torre

In agent communication languages, the inferences that can be made on the basis of a communicative action are inherently conditional, and non-monotonic. For example, a proposal only leads to a commitment, on the condition that it is accepted. And in a persuasion dialogue, assertions may later be retracted. In this paper we therefore present a defeasible logic that can be used to express a semantics for agent communication languages, and to efficiently make inferences on the basis of communicative actions. The logic is non-monotonic, allows nested rules and mental attitudes as the content of communicative actions, and has an explicit way of expressing persistence over time. Moreover, it expresses that mental attitudes are publicly attributed to agents playing roles in the dialogue. To illustrate the usefulness of the logic, we reformalize the meta-theory underlying the FIPA semantics for agent communication, focusing on inform and propose. We show how composed speech acts can be formalized, and extend the semantics with an account of persuasion.

nrac07.pdf

Contextual Deliberation of Cognitive Agents in Defeasible Logic

Guido Governatori, Mehdi Dastani, Antonino Rotolo, Insu Song, Leendert van der Torre

Relationships Meet their Roles in Object Oriented Programming

Matteo Baldoni, Guido Boella and Leendert van der Torre

In this paper we study how roles can be added to patterns modelling relationships in Object Oriented programming. Relationships can be introduced in programming languages either by reducing them to attributes of the objects which participate in the relationship, or by modelling the relationship itself as a class whose instances have the participants of the relationships among their attributes. However, even if roles have been recognized as an essential component of relationships, also in modelling languages like UML, they have not been introduced in Object Oriented programming when it is necessary to model relationships. Introducing roles allows to add attributes and behaviors to the participants in the relationship, rather than to the relationship itself, and to distinguish the natural types of the participants in the relationships from the roles the participants acquire in the relationships. We show how the role model of the language powerJava can be used to endow the relationship as attribute pattern with roles.

fsem07.pdf

I fondamenti ontologici dei linguaggi di programmazione orientati agli oggetti: i casi delle relazioni e dei ruoli

Matteo Baldoni, Guido Boella and Leendert van der Torre

The Ontological Properties of Social Roles in Multi-agent Systems: Definitional Dependence, Powers and Roles Playing Roles

Guido Boella and Leendert van der Torre

Institutions with a Hierarchy of Authorities in Distributed Dynamic Environments

Guido Boella and Leendert van der Torre

A single global authority is not sufficient to regulate heterogenous agents in multiagent systems based on distributed architectures, due to idiosyncratic local situations and to the need to regulate new issues as soon as they arise. On the one hand institutions should be structured as normative systems with a hierarchy of authorities able to cope with the dynamics of local situations, but on the other hand higher authorities should be able to delimit the autonomy of lower authorities to issue valid norms. In this paper we study the interplay of obligations and strong permissions in the context of hierarchies of authorities using input/output logic, because its explicit norm base facilitates reasoning about norm base maintenance, and it covers a variety of conditional obligations and permissions. We combine the logic with constraints, priorities and hierarchies of authorities. In this setting we observe that Makinson and van der Torre's notion of prohibition immunity for permissions is no longer sufficient, and we introduce a new notion of permission as exception and a new distinction between static and dynamic norms. We show how strong permissions can dynamically change an institution by adding exceptions to obligations, provide an explicit representation of what is permitted to the subjects of the normative system and allow higher level authorities to limit the power of lower level authorities to change the normative system.

ialaw07a.pdf

The Roles of Roles in Agent Communication Languages

Guido Boella, Rossana Damiano, Joris Hulstijn, and Leendert van der Torre

We consider agents having multiple communication sessions at the same time. We assume that FIPA semantics of agent communication languages can still be used when we attribute mental attitudes for each session, which we call the roles of the agents, and we assume that we have to distinguish the mental attitudes attributed to the roles from the mental attitudes of the agents. We consider several consequences of the distinction between the mental attitudes attributed to the roles and the mental attitudes attributed to the agent. First, in attributing mental attitudes to an agent or to one of its roles, we argue that only mental attributes are attributed to an agent's role when these attributes follow directly from the agent's communication. They are therefore public in the sense that every agent who has overheard the session, has the same beliefs about the mental attitudes of the role. Second, the moves permitted to the dialogue participants in the same dialogue game are based on the role only, such that different kind of moves can be specified in different types of dialogue games. Obligations are associated to roles related to institutions which can enforce them by means of sanctions. Third, expectations are based both on the mental attitudes ascribed to the agent and to the role.

iat06.pdf

Distinguishing Propositional and Action Commitment in Agent Communication

Guido Boella, Rossana Damiano, Joris Hulstijn, and Leendert van der Torre

Our goal is to extend agent communication languages for persuasion dialogues. We distinguish action commitments from propositional commitments, because both limit future moves, but an action commitment is fulfilled when the hearer believes that the action is performed, whereas a propositional commitment is fulfilled only when the hearer concedes to the proposition - where concessions are the absence of a belief to the contrary, and prevent further challenges. Using a common model for both kind of commitments and a role-based semantics of agent communication languages, we show how propositional commitments are related to public beliefs and action commitments to public goals.

cmna06.pdf

From Social Power to Social Importance

Guido Boella, Luigi Sauro and Leendert van der Torre

Interaction among Objects via Roles

Matteo Baldoni, Guido Boella and Leendert van der Torre

In this paper we present a new vision in object oriented programming languages where the objects' attributes and operations depend on who is interacting with them. This vision is based on a new definition of the notion of role, which is inspired to the concept of affordance as developed in cognitive science. The current vision of objects considers attributes and operations as being objective and independent from the interaction. In contrast, in our model interaction with an object always passes through a role played by another object manipulating it. The advantage is that roles allow to define operations whose behavior changes depending on the role and the requirements it imposes, and to define session aware interaction, where the role maintains the state of the interaction with an object. Finally, we discuss how roles as affordances can be introduced in Java, building on our language powerJava.

pppj06.pdf, pppj06.ppt

Introduction to Normative Multiagent Systems

Guido Boella, Leendert van der Torre and Harko Verhagen

This article introduces the research issues related to and definition of normative multiagent systems. It also describes the papers selected from NorMAS05 that are part of this double special issue and relates the papers to each other.

cmot06.pdf

Norm Negotiation Power

Guido Boella and Leendert van der Torre

In social mechanism design, norm negotiation creates individual or contractual obligations fulfilling goals of the agents. The social delegation cycle distinguishes among social goal negotiation, obligation and sanction negotiation and norm acceptance. Power may affect norm negotiation in various ways, and we therefore introduce a new formalization of the social delegation cycle based on power and dependence, without referring to the rule structure of norms, actions, decision variables, tasks, and so on.

bnaic06.pdf

Game-Theoretic Foundations for Norms

Guido Boella and Leendert van der Torre

In this paper we study game-theoretic foundations for norms. We assume that a norm is a mechanism to obtain desired multi-agent system behavior, and must therefore under normal or typical circumstances be fulfilled by a range of agent types, such as norm internalizing agents, respectful agents fulfilling norms if possible, and selfish agents obeying norms only due to the associated sanctions.

polishai06.pdf polishai06-full.pdf

Importing Agent-like Interaction in Object Orientation

Matteo Baldoni, Guido Boella and Leendert van der Torre

This paper begins with the comparison of the message-sending mechanism, for communication among agents, and the method-invocation mechanism, for communication among objects. Then, we describe an extension of the methodinvocation mechanism by introducing the notion of "sender" of a message, "state" of the interaction and "protocol" using the notion of "role", as it has been introduced in the powerJava extension of Java. The use of roles in communication is shown by means of an example of protocol.

woa06.pdf woa06.ppt

Norm Negotiation in Multiagent Systems

Guido Boella and Leendert van der Torre

Normative multiagent systems provide agents with abilities to autonomously devise societies and organizations coordinating their behavior via social norms and laws. In this paper we study how agents negotiate new social norms and when they accept them. We introduce a negotiation model based on what we call the social delegation cycle, which explains the negotiation of new social norms from agent desires in three steps. First individual agents or their representatives negotiate social goals, then a social goal is negotiated in a social norm, and finally the social norm is accepted by the agents when it leads to fulfilment of the desires the cycle started with. We characterize the allowed proposals during social goal negotiation as mergers of the individual agent desires, and we characterize the allowed proposals during norm negotiation as both joint plans to achieve the social goal (obligations associated with the norm) and the associated sanctions or rewards (a control system associated with the norm). The norm is accepted when the norm is stable in the sense that agents will act according to the norm, and effective in the sense that fulfilment of the norm leads to achievement of the agents' desires. We also compare norm negotiation with contract negotiation and negotiation of the distribution of obligations.

ijcis06.pdf

Modelling The Interaction Between Objects: Roles as Affordances

Matteo Baldoni, Guido Boella and Leendert van der Torre

In this paper we present a new vision of objects in ontologies where the objects' attributes and operations depend on who is interacting with them. This vision is based on a new definition of the notion of role, which is inspired by the concept of affordance as developed in cognitive science. The current vision of objects considers attributes and operations as being objective and independent from the interaction. In contrast, in our model interaction with an object always passes through a role played by another object manipulating it. The advantage is that roles allow to define operations whose behavior changes depending on the role and the requirements it imposes, and to define session aware interaction, where the role maintains the state of the interaction with an object. Finally, we provide a description of the model in UML and we discuss how roles as affordances have been introduced in Java.

ksem06.pdf

Merging Rules: Preliminary Version

Richard Booth, Souhila Kaci, and Leendert van der Torre

In this paper we consider the merging of rules or conditionals. In contrast to other approaches, we do not invent a new approach from scratch, for one particular kind of rule, but we are interested in ways to generalize existing revision and merging operators from belief merging to rule merging. First, we study ways to merge rules based on only a notion of consistency of a set of rules, and illustrate this approach using a consolidation operator of Booth and Richter. Second, we consider ways to merge rules based on a notion of implication among rules, and we illustrate this approach using socalled min and max merging operators defined using possibilistic logic.

nmr06c.pdf

Acyclic Argumentation: Attack = Conflict + Preference

Souhila Kaci, Leendert van der Torre and Emil Weydert

In this paper we study the fragment of Dung's argumentation theory in which the strict attack relation is acyclic. We show that every attack relation satisfying a particular property can be represented by a symmetric conflict relation and a transitive preference relation in the following way.We define an instance of Dung's abstract argumentation theory, in which `argument A attacks argument B' is defined as `argument A conflicts with argument B' and `argument A is at least as preferred as argument B', where the conflict relation is symmetric and the preference relation is transitive.We show that this new preference-based argumentation theory characterizes the acyclic strict attack relation, in the sense that every attack relation defined as such a combination satisfies the property, and for every attack relation satisfying the property we can find a symmetric conflict relation and a transitive preference relation satisfying the equation.

ecai06d.pdf

Count-as Conditionals, Classification and Context

Guido Boella and Leendert van der Torre

Searle represents constitutive norms as count-as conditionals, written as `X counts as Y in context C'. Grossi et al. study a class of these conditionals as `in context C, X is classified as Y'. In this paper we propose a generalization of this relation among count-as conditionals, classification and context, by defining a class of count-as conditionals as `X in context C0 is classified as Y in context C'. We show that if context C0 can be different from context C, then we can represent a larger class of examples, and we have a weaker logic of count-as conditionals.

ecai06c.pdf

Fair Distribution of Collective Obligations

Guido Boella and Leendert van der Torre

In social mechanism design, obligation distribution creates individual or contractual obligations that imply a collective obligation. A distinguishing feature from group planning is that also the sanction of the collective obligation has to be distributed, for example by creating sanctions for the individual or contractual obligations. In this paper we address fairness in obligation distribution for more or less powerful agents, in the sense that some agents can perform more or less actions than others. Based on this power to perform actions, we characterize a trade-off in negotiation power. On the one hand, more powerful agents may have a disadvantage during the negotiation, as they may be one of the few or even the only agent who can see to some of the actions that have to be performed to fulfill the collective obligation. On the other hand, powerful agents may have an advantage in some negotiation protocols, as they have a larger variety of proposals to choose from. Moreover, powerful agents have an advantage because they can choose from a larger set of possible coalitions. We present an ontology and measures to find a fair tradeoff between these two forces in social mechanism design.

ecai06b.pdf

Strengthening Admissible Coalitions

Guido Boella, Luigi sauro and Leendert van der Torre

We develop a criterion for coalition formation among goal-directed agents, the indecomposable do-ut-des property. The indecomposable do-ut-des property refines the do-ut-des property (literally give to get) by considering the fact that agents prefer to form coalitions whose components cannot be formed independently. A formal description of this property is provided as well as an analysis of algorithms and their complexity.

ecai06a.pdf

Preference reasoning for argumentation: non-monotonicity and algorithms

Souhila Kaci and Leendert van der Torre

In this paper we are interested in the role of preferences in argumentation theory. To promote a higher impact of preference reasoning in argumentation, we introduce a novel preference-based argumentation theory. Using non-monotonic preference reasoning we derive a Dung-style attack relation from a preference specification together with a defeat relation. In particular, our theory uses efficient algorithms computing acceptable arguments via a unique preference relation among arguments from a preference relation among sets of arguments.

nmr06a.pdf

Merging Optimistic and Pessimistic Preferences

Souhila Kaci and Leendert van der Torre

In this paper we consider the extension of nonmonotonic preference logic with the distinction between controllable (or endogenous) and uncontrollable (or exogenous) variables, which can be used for example in agent decision making and deliberation. We assume that the agent is optimistic about its own controllables and pessimistic about its uncontrollables, and we study ways to merge these two distinct dimensions. We also consider complex preferences, such as optimistic preferences conditional on an uncontrollable, or optimistic preferences conditional on a pessimistic preference.

nmr06b.pdf

A logical architecture of a normative system

Guido Boella and Leendert van der Torre

Logical architectures combine several logics into a more complex logical system. In this paper we study a logical architecture using input/output operations corresponding to the functionality of logical components. We illustrate how the architectural approach can be used to develop a logic of a normative system based on logics of counts-as conditionals, institutional constraints, obligations and permissions. In this example we adapt for counts-as conditionals and institutional constraints a proposal of Jones and Sergot, and for obligations and permissions we adapt the input/output logic framework of Makinson and van der Torre. We use our architecture to study logical relations among counts-as conditionals, institutional constraints, obligations and permissions. We show that in our logical architecture the combined system of counts-as conditionals and institutional constraints reduces to the logic of institutional constraints, which again reduces to an expression in the underlying base logic. Counts-as conditionals and institutional constraints are defined as a pre-processing step for the regulative norms. Permissions are defined as exceptions to obligations and their interaction is characterized.

deon06c.pdf

Delegation of Power in Normative Multiagent Systems

Guido Boella and Leendert van der Torre

In this paper we reconsider the definition of counts-as relations in normative multiagent systems: counts-as relations do not always provide directly an abstract interpretation of brute facts in terms of institutional facts. We argue that in many cases the inference of institutional facts from brute facts is the result of actions of agents acting on behalf of the normative systems and who are in charge of recognizing which institutional facts follow from brute facts. We call this relation delegation of power: it is composed of a counts-as relation specifying that the effect of an action of an agent is an institutional fact and by a goal of the normative system that the fact is considered as an institutional fact. This relation is more complex than institutional empowerment, where an action of an agent counts-as an action of the normative system but no goal is involved, and than delegation of goals, where a goal is delegated to an agent without giving it any power.With two case studies we show the importance of the delegation of power. Finally, we show how the new definition can be related with existing ones by using different levels of abstraction.

deon06b.pdf

Permissions and Uncontrollable Propositions in DSDL3: Non-Monotonicity and Algorithms

Souhila Kaci and Leendert van der Torre

In this paper we are interested in non-monotonic extensions of Bengt Hansson's standard dyadic deontic logic 3, known as DSDL3. We study specificity principles for DSDL3 with both controllable and uncontrollable propositions. We introduce an algorithm for minimal specificity which not only covers obligations but also permissions, and we discuss the distinction between weak and strong permissions. Moreover, we introduce ways to combine algorithms for minimal and maximal specificity for DSDL3 with controllable and uncontrollable propositions, based on 'optimistic' and 'pessimistic' reasoning respectively.

deon06a.pdf

ACL Semantics between Social Commitments and Mental Attitudes

Guido Boella, Rosanna Damiano, Joris Hulstijn and Leendert van der Torre

There are two main traditions in defining a semantics for agent communication languages, based either on mental attitudes or on social commitments. In this paper, we translate both traditions in a different approach in which the dialogue state is represented by the beliefs and goals publicly attributed to the roles played by the dialogue participants. On the one hand, this approach avoids the problems of mentalistic semantics, such as the unverifiability of private mental states. On the other hand, it can be applied also to cooperative situations that can be modelled by using a weaker notion than commitment.

ac06.pdf http://dx.doi.org/10.1007/978-3-540-68143-4_3

Coordination and Organization: Definitions, Examples and Future Research Directions

Guido Boella and Leendert van der Torre

Coordination languages and models like Linda and Reo have been developed in com- puter science to coordinate the interaction among components and objects, and are nowadays used to model and analyze organizations too. Moreover, organizational concepts are used to enrich the existing coordination languages and models. We describe this research area of \organization and coordination" by presenting de- nitions, examples, and future research directions. We highlight two issues. First, we argue for a study of value-based rather than information-based coordination languages to model the coordination of autonomous agents and organizations. Sec- ond, we argue for a study of the balance between enforced control and trust-based anticipation to deal with security aspects in the coordination of organizations.

entcs06a.pdf

A Foundational Ontology of Organizations and Roles

Guido Boella and Leendert van der Torre

In this paper we propose a foundational ontology of the social concepts of organization and role which structure institutions. We identify which axioms model social concepts like organization and role and which properties distinguish them from other categories like objects and agents: the organizational structure of institutions, the relation between roles and organizations, and the powers among the components of an organization. All social concepts depend on descriptions defining them, which are collectively accepted, and the description defining the components of organizations, including roles, are included in the description of the organizations they belong to. Thus, the relational dependence of roles means that they are defined in the organizations they belong to. Finally, powers inside organizations are defined by the fact that components of an organization can access the state of the organization whose definition they depend on and of the other components, thus violating the standard encapsulation principle of objects.

dalt06.pdf

Bridging Agent Theory and Object Orientation: Agent-like Communication among Objects

Matteo Baldoni, Guido Boella and Leendert van der Torre

This paper begins with the comparison of the message-sending mechanism, for communication among agents, and the method-invocation mechanism, for communication among objects. Then, we describe an extension of the method-invocation mechanism by introducing the notion of sender" of a message, state" of the interaction and protocol" using the notion of role", as it has been introduced in the powerJava extension of Java. The use of roles in communication is shown by means of an example of protocol.

promas06.pdf

Landscape maps for enterprise architectures

Leendert van der Torre, Marc Lankhorst, Hugo ter Doest, Jan Campschroer, Farhad Arbab

Landscape maps are a technique for visualizing enterprise architectures. They present architectural elements in the form of an easy to understand 2D map. A landscape map view on architectures provides non-technical stakeholders, such as managers, with a high-level overview, without burdening them with technicalities of architectural drawings. In this paper we discuss the use of and techniques for landscape maps. A formal model for landscape maps is introduced as the basis of visualization and interaction techniques. Moreover, we show how a landscape map can be generated from its underlying model. Finally we show several interaction techniques, for example to build a landscape map from scratch, independently of an underlying model, or to change a landscape map together with its underlying model.

caise06.pdf

Role-based semantics for agent communication: Embedding of the Mental Attitudes and Social Commitments Semantics

Guido Boella, Joris Hulstijn, Rossana Damiano and Leendert van der Torre

In this paper we illustrate how a role-based semantics for agent communication languages can embed the two main existing models of agent communication languages, respectively based on mental attitudes and social commitments semantics. These two models have been presented as incompatible approaches, but recently we illustrated for persuasion dialogues and using our normative multiagent systems framework, that they can be seen also as complimentary ones. Independently from our own multi-agent model, in this paper we illustrate for the speech act inform how the role based semantics embeds the other two semantics.

aamas06b.pdf

An architecture of a normative system

Guido Boella and Leendert van der Torre

Normative systems are traditionally described and analyzed using deontic logic, describing the logical relations among obligations and permissions. However, there is still a gap between deontic logic and normative multi-agent systems such as electronic institutions, which may be seen as an instance of the gap between on the one hand logical agent specification languages and on the other hand agent architectures and programming languages. To bridge the gap, in this paper we propose an architecture containing separate subsystems or components for counts-as conditionals, conditional obligations and conditional permissions. We add a norm database component in which the three kinds of rules are stored, and we use a channel based coordination model to describe the relations among the four normative components.

aamas06a.pdf

Powerjava: ontologically founded roles in object oriented programming languages

Matteo Baldoni, Guido Boella and Leendert van der Torre

sac06.pdf

In this paper we introduce a new view on roles in Object Oriented programming languages, based on an ontological analysis of roles. A role is always associated with an object instance playing the role and also to an object instance (its institution) which represents its context. The denition of a role depends on the denition of the institution. This property allows to endow role-players with powers that can modify the state of the institution and of the other roles dened in it. As an example, we introduce a role construct in Java, where the abolve features are interpreted as follows. Roles are implemented as classes, which can be instantiated only in presence of an instance of the player and of an instance of an institution. The denition of a class implementing a role is included in the class of the institution, the role belongs to. Powers are methods which can access private elds and methods of the institution they belong to, and of the other roles of the same institution.

Social roles, from agents back to agents

Matteo Baldoni, Guido Boella and Leendert van der Torre

In this paper we introduce a new view on roles in Object Oriented programming languages. This view is based on an ontological analysis of roles and attributes to roles the following properties: first, a role is always associated not only with an object instance playing the role, but also to another object instance which constitutes the context of the role and which we call institution. Second, the definition of a role depends on the definition of the institution which constitutes its context. Third, this second property allows to endow players of roles with powers to modify the state of the institution and of the other roles of the same institution. As an example of this model of roles in Object Oriented programming languages, we introduce a role construct in Java. We interpret these three features of roles in Java as the fact that, first, roles are implemented as classes which can be instantiated only in presence of an instance of the player of the role and of an instance of the class representing the institution. Second, the definition of a class implementing a role is included in the class definition of the institution the role belongs to. Thirdly, powers are methods of roles which can access private fields and methods of the institution they belong to and of the other roles of the same institution.

woa05.pdf

Introducing Ontologically Founded Roles in Object Oriented Programming: powerJava

Matteo Baldoni, Guido Boella and Leendert van der Torre

In this paper we introduce a new view on roles in Object Oriented programming languages. Our notion is based on an ontological analysis of social roles and attributes to roles the following properties: first, a role is always associated not only to an object instance playing the role, but also to another object instance which constitutes the context of the role and which we call institution. Second, the definition of a role depends on the definition of the institution which constitutes its context. Third, this second property allows to endow players of roles with powers to modify the state of the institution and of the other roles of the same institution. As an example of this model of roles in Object Oriented programming languages, we introduce a role construct in Java.

roles05.pdf

Programming Cognitive Agents in Defeasible Logic

Mehdi Dastani, Guido Governatori, Antonino Rotolo, Leendert van der Torre

Defeasible Logic is extended to programming languages for cognitive agents with preferences and actions for planning. We define rule-based agent theories that contain preferences and actions, together with inference procedures.We discuss patterns of agent types in this setting. Finally, we illustrate the language by an example of an agent reasoning about web-services.

lpar05.pdf

Semantic Analysis of Chisholm's Paradox

Jan Broersen and Leendert van der Torre

Violation handling is a crucial problem in many applications. Therefore its paradoxes have been studied in, amongst others, artificial intelligence, agent theory and computer science. The standard way to study these paradoxes is to model them using a formal language, and use formal logic to consider whether the set of sentences is inconsistent, the sentences logically follow from others, or some other anomaly occurs. During the past decades, developments in temporal, action and non-monotonic logics have contributed to a better understanding of the paradoxes and thus of violation handling. In this paper we propose an alternative way to analyze Chisholm's notorious contrary-to-duty paradox in deontic logic. We model the paradox using semantic models, using insights from conceptual modelling. We aim to gain insight in the open question whether the paradoxes are in some sense logical contradictions, or only apparent contradictions. If a paradox is only an apparent contradiction, then there has to be a model interpreting all sentences.

bnaic05.pdf

Preferences of Agents in Defeasible Logic

Mehdi Dastani, Guido Governatori, Antonino Rotolo, Leendert van der Torre

We are interested in programming languages for cognitive agents with preferences. We define rule-based agent theories and inference procedures in defeasible logic, and in this setting we discuss patterns of agent behavior called agent types.

ai05.pdf

Security Policies for Sharing Knowledge in Virtual Communities

Guido Boella and Leendert van der Torre

Knowledge management exploits the new opportunities of sharing knowledge among members of virtual communities in distributed computer networks, and knowledge-management systems are therefore modeled and designed as multiagent systems. In this paper, normative multiagent systems for secure knowledge management based on access-control policies are studied. It is shown how distributed access control is realized by means of local policies of access-control systems for documents of knowledge providers, and by means of global community policies regulating these local policies. Moreover, it is shown how such a virtual community of multiple knowledge providers respects the autonomy of the knowledge providers.

smca06.pdf

Formalisation and Analysis of the Temporal Dynamics of Conditioning

Tibor Bosse, Catholijn M. Jonker, Sander A. Los, Leendert van der Torre and Jan Treur

Coordination and Organization 05 (event report)

Guido Boella and Leendert van der Torre

Organizations embody a powerful way to coordinate complex behavior in human society. Different models of organisations exist, from bureaucratic systems based on norms to competitive systems based on markets. Moreover, organizational concepts allow to structure the behavior of complex entities in a hierarchy of encapsulated entities: departments structured in roles, organisations structured in departments, and interorganizational coordination structured in organizations. Organizations specify the interaction and communication possibilities of each of these entities, abstracting from the implementation of their behavior. Since these entities are autonomous, they can only be coordinated exogeneously.

agentlink05.pdf

Argument games for interactive access control

Guido Boella, Joris Hulstijn, and Leendert van der Torre

We are interested in interactive access control to web services in virtual organizations. We discuss argument games in which the set of credentials requested by the service provider to access a service is established by means of an interaction between a client acting as a proponent and a server acting as an opponent.

wi05.pdf

Interaction in Normative Multiagent Systems

Guido Boella, Joris Hulstijn, and Leendert van der Torre

The central research question of this paper is how notions developed in interactive computing such as abstract behavior types, the coordination language Reo, and Boolean circuits with registers, can be used to extend logical input/output nets, or lions for short. Lions are based on input/output logic, a deontic logic which is not used as a (non-classical) inference engine deriving output from input, but as a secretarial assistant for logically assisted transformations from input to output. We consider two extensions of input/output logics and lions. First, we consider input/output logics defined on infinite sequences (or streams) of inputs and outputs. Secondly, we consider lions with AND and register gates, formalizing the behavior of channels and connectors. We discuss also the role of interactive computing in normative multi-agent systems motivating the development of lions.

entcs05.pdf

A Synthesis Between Mental Attitudes and Social Commitments in Agent Communication Languages

Guido Boella, Joris Hulstijn, and Leendert van der Torre

There are two main traditions in defining a semantics for agent communication languages, based either on mental attitudes or on social commitments. In this paper we show how the role metaphor can be used to bridge the gap between these two approaches. First, we show how dialogues can be modelled as games - a form of normative systems - and how mental attitudes can be attributed not only to agents, but also, in a public manner, to the roles of the game. The dialogue moves allow an agent playing a role to modify the roles' mental states, as specified by the counts-as conditionals (also known as constitutive norms) defining the game. The player of a role is expected to act as if it has the mental attitudes attributed to its role during the dialogue and to prevent its role's mental attitudes from becoming incoherent, as it does for its own private mental attitudes. Secondly, we show how roles as descriptions of expected behavior maintain the normative character of social semantics. Due to the bridge between the two approaches, results and tools from one approach can be used in the other one.

iat05.pdf

Admissible Agreements among Goal-directed Agents

Guido Boella, Luigi Sauro, and Leendert van der Torre

We study admissible coalitions in goal-directed multiagent systems.We define a qualitative criterion of admissibility in which a coalition has itself all the necessary information to check admissibility. We show also that, under some assumptions on preference relations of the agents, this admissibility criterion can be used to reduce the search space in a game theoretical approach.

iat05b.pdf

Role-based Rights in Artificial Social Systems

Guido Boella and Leendert van der Torre

In this paper we use normative systems to introduce roles and rights in the game-theoretic artificial social systems developed by Shoham and Tennenholtz. We model normative systems as socially constructed agents whose behavior is determined by a set of role playing agents. Roles are again modeled as socially constructed agents, and the roles' behavior is the ideal behavior of agents playing the roles. In our approach, the strategies of the role correspond to the rights that can be exercised by the role. In other words, rights are powers extending the set of strategies of an agent - not constraining them! - due to the new opportunities to exercise rights. We consider the role assignment problem of how to assign agents to roles such that the role playing agent is expected to behave like the ideal behavior of the role. We also consider how the normative system controls the behavior of agents playing a role in it.

iat05a.pdf

Argumentation for access control

Guido Boella, Joris Hulstijn, and Leendert van der Torre

In this paper we are interested in argument based reasoning for access control, for example in the context of agents negotiating access to resources or web services in virtual organizations.We use a logical framework which contains agents with objectives concerning access to a resource or provision of a service, including security objectives. The access control mechanism is described by a set of policy rules, that specify that access to a resource or service requires a specific set of credentials. Our contribution is a formalization of the reasoning about access control using a planning theory formalized in Dung's abstract argumentation framework.We build on Amgoud's argumentation framework for plan arguments, which is based on an adaptation of Dung's notion of defence. Our formal argumentation framework allows arguments about the backward derivation of plans from objectives and policy rules (abduction), as well as arguments about the forward derivation of goals from general objectives. We show that reasoning about the feasibility of goals requires mixed goal-plan arguments, and we show how to formalize the plan arguments in Dung's framework without adapting the notion of defence.

aiia05.pdf

Change Impact Analysis of Enterprise Architectures

F. de Boer, M. Bonsangue, L. Groenewegen, A. Stam, S. Stevens, and L. van der Torre

An enterprise architecture is a high-level description intended to capture the vision of an enterprise integrating all its dimensions: organization structure, business processes, and infrastructure. Every single part of an enterprise is subject to change, and each change may have significant consequences within all domains of the enterprise. A lot of effort is therefore devoted on maintaining the integrity of an architectural description.

In this paper we address the problem of mastering the ripple effects of a proposed change. This allows architects to assess the consequences of a particular change to the enterprise, so to identify potential impacts of a change before it actually takes place.

iri05.pdf

Enterprise architecture at work

Marc Lankhorst et al

An enterprise architecture tries to describe and control an organisation's structure, processes, applications, systems and techniques in an integrated way. The unambiguous specification and description of components and their relationships in such an architecture requires a coherent architecture modelling language.

Lankhorst and his co-authors present such an enterprise modelling language that captures the complexity of architectural domains and their relations and allows the construction of integrated enterprise architecture models. They provide architects with concrete instruments that improve their architectural practice. As this is not enough, they additionally present techniques and heuristics for communicating with all relevant stakeholders about these architectures. Since an architecture model is useful not only for providing insight into the current or future situation but can also be used to evaluate the transition from `as-is' to `to-be', the authors also describe analysis methods for assessing both the qualitative impact of changes to an architecture and the quantitative aspects of architectures, such as performance and cost issues.

The modelling language and the other techniques presented have been proven in practice in many real-life case studies. So this book is an ideal companion for enterprise IT or business architects in industry as well as for computer or management science students studying the field of enterprise architecture.

Constitutive Norms in the Design of Normative Multiagent Systems

Guido Boella and Leendert van der Torre

In this paper, we consider the design of normative multiagent systems composed of both constitutive and regulative norms. We analyze the properties of constitutive norms, in particular their lack of reflexivity, and the trade-off between constitutive and regulative norms in the design of normative systems. As methodology we use the metaphor of describing social entities as agents and of attributing them mental attitudes. In this agent metaphor, regulative norms expressing obligations and permissions are modelled as goals of social entities, and constitutive norms expressing "counts-as" relations are their beliefs.

clima05.pdf

The Ontological Properties of Social Roles: Definitional Dependence, Powers and Roles Playing Roles

Guido Boella and Leendert van der Torre

In this paper we address the problem of defining social roles in MAS. Social roles provide the basic structure of social institutions and organizations. We start from the properties attributed to roles both in the MAS and the Object Oriented community, and we use them in an ontological analysis of the notion of social role. We thus identify as the main properties of social roles being definitionally dependent on the institution they belong to, attributing powers to the agents playing them, and allowing roles to play roles. The methodology we use to model roles is the agent metaphor: social roles, in the same way as social institutions, like normative systems and organizations, are attributed mental attitudes to explain their behavior.

loait05.doc

Non-monotonic reasoning with various kinds of preferences

Souhila Kaci and Leendert van der Torre

We are interested in systems which do not prescribe one single kind of preference, but in which varying kinds of preferences can be used simultaneously. In such systems it is essential to know the interaction among the kinds of preferences being used, and we therefore introduce and study a nonmonotonic logic to reason about sixteen strict and non-strict kinds of preferences, including ceteris paribus preferences. Moreover, we study "distinguished" preference orders based on specificity principles by showing when these distinguished preference orders are unique, and by presenting algorithms to calculate the distinguished preference orders.

pref05.pdf

Transaction trust in normative multiagent systems

Guido Boella, Joris Hulstijn, Yao-Hua Tan, and Leendert van der Torre

In this paper we apply Boella and van der Torre's normative multiagent systems to analyze a model of transaction trust introduced by Tan and colleagues. We discuss the role of constitutive and regulative norms in party trust and control trust. Moreover, we use so called contract negotiation games to indicate when controls are needed. Though sometimes regulators focus exclusively on adding controls, a scenario analysis can be used to show when this is not needed. We illustrate these issues of transaction trust by a case study from international trade, namely the Letter of Credit procedure.

trust05.pdf

From the Theory of Mind to the Construction of Social Reality

Guido Boella and Leendert van der Torre

In this paper we argue that the hypothesis of the theory of mind advanced in cognitive science can be the basis not only of the social abilities which allow interaction among individuals, but also of the construction of social reality. The theory of mind is the attribution, via the agent metaphor, of mental attitudes, like beliefs and goals, to other agents. Analogously, we attribute mental attitudes to social entities, like groups, normative systems and organizations with roles. The agent metaphor explains the necessary abilities to deal with complex aspects of social behavior, like acting in a group, playing a role in an organization, and living in a reality organized in institutions which create regulative and constitutive norms to regulate behavior. To show the feasibility of this approach we provide a computational model of the construction of social reality based on multiagent systems.

cogsci05.pdf

Design By Contract Deontic Design Language for Multiagent Systems

Christophe Garion and Leendert van der Torre

Design by contract is a well known theory that views software construction as based on contracts between clients (callers) and suppliers (routines), relying on mutual obligations and benefits made explicit by assertions. However, there is a gap between this theory and software engineering concepts and tools. For example, dealing with contract violations is realized by exception handlers, whereas it has been observed in the area of deontic logic in computer science that violations and exceptions are distinct concepts that should not be confused. To bridge this gap, we propose a software design language based on temporal deontic logic. Moreover, we show how preferences over the possible outcomes of a supplier can be added. We also discuss the relation between the normative stance toward systems implicit in the design by contract approach and the intentional or BDI stance popular in agent theory.

anirem05a.pdf

Applying Normative Multiagent Systems: A Case Study

Guido Boella, Joris Hulstijn, Yao-Hua Tan, and Leendert van der Torre

We present a conceptual model of control mechanisms for virtual organizations. Control mechanisms and norms in general have certain objectives. We propose to model both the participants, and the normative system itself as autonomous agents, having certain beliefs, desires and goals. Norms, which can be internalized by the agents as obligations, are translated into conditional beliefs, desires and goals of the normative system, which concern both detection and sanctioning measures. The model is illustrated by a case study of the Renewable Obligation Certificates (ROC), introduced in the United Kingdom to stimulate the production of renewable energy. We show that the model can handle both the regulative and the evidential aspects of the case.

anirem05b.pdf

Organizations in Artificial Social Systems

Guido Boella and Leendert van der Torre

In this paper we introduce organizations and roles in Shoham and Tennenholtz' artificial social systems, using a normative system. We model how real agents determine the behavior of organizations by playing roles in the organization, and how the organization controls the behavior of agents playing a role in it. We consider the design of an organization in terms of roles and the assignment of agents to roles, and the evolution of organizations. We do not present a complete formalization of the computational problems, but we illustrate our approach by examples.

ooop05.pdf

A logic of abstract argumentation

Guido Boella, Joris Hulstijn and Leendert van der Torre

In this paper we introduce a logic of abstract argumentation capturing Dung's theory of abstract argumentation, based on connectives for attack and defend. We extend it to a modal logic of abstract argumentation to generalize Dung's theory and define variants of it. Moreover, we use the logic to relate Dung's theory of abstract argumentation to more traditional conditional and comparative formalisms, and we illustrate how to reason about arguments in meta-argumentation.

argmas05.pdf

Formalisation and Analysis of the Temporal Dynamics of Conditioning

Tibor Bosse, Catholijn Jonker, Sander Los, Leendert van der Torre and Jan Treur

In order to create adaptive Agent Systems with abilities matching those of their biological counterparts, a natural approach is to incorporate classical conditioning mechanisms into such systems. However, existing models for classical conditioning are usually based on differential equations. Since the design of Agent Systems is traditionally based on qualitative conceptual languages, these differential equations are often not directly appropriate to serve as an input for Agent System design. To deal with this problem, this paper explores a formal description and analysis of a conditioning process based on logical specification and analysis methods of dynamic properties of conditioning. Specific types of dynamic properties are global properties, describing properties of the process as a whole, or local properties, describing properties of basic steps in a conditioning process. If the latter type of properties are specified in an executable format, they provide a temporal declarative specification of a simulation model. Global properties can be checked automatically for simulated or other traces. Using these methods the properties of conditioning processes informally expressed by Los and Heuvel [7] have been formalised and verified against a specification of local properties based on Machado [8]'s mathematical model.

aose05.pdf

Bridging Agent Theory and Object Orientation: Importing Social Roles in Object Oriented Languages

Matteo Baldoni, Guido Boella, and Leendert van der Torre

Social roles structure social institutions like organizations in Multi- Agent Systems (MAS). In this paper we describe how to introduce the notion of social role in programming languages. To avoid the commitment to a particular agent model, architecture or language, we decided to extend Java, the most prominent object oriented programming language, by adding social roles. The obtained language allows an easier implementation of MAS's w.r.t. the Java language. We also show that many important properties of social roles, studied in the MAS field, can be applied to objects. Two are the essential features of social roles according to an analysis reported in the paper: social roles are defined by other entities (called institutions), and when an agent plays a role it is endowed with powers by the institution that defines it. We interpret these two features into the object oriented paradigm as the fact that social roles are objects, which are defined in and exist only inside other objects (corresponding to institutions), and that, through a role, external objects playing the role can access to the object (institution) the role belongs to.

promas05.pdf

Permission and authorization in Normative Multiagent Systems

Guido Boella, and Leendert van der Torre

The distinction between the notions of permission and authorization is subtle. In the Cambridge Advanced Learner's Dictionary [4] permitting is "to allow something", "to make it possible for someone to do something, or to not prevent something from happening", while authorizing means "to give someone official permission to do something". Law studies argue that the distinction goes beyond the "officiality" of authorization. E.g., the Del Giudice [5]'s dictionary of law argues that adding or removing an authorization does not change the normative status of an agent while adding or removing a permission does. Authorizations change what is obligatory or permitted for agents without adding or removing norms. However, though legal philosophers distinguish permission from authorization, the distinction between the two is ignored in many (agent) theories and systems. How can this apparent paradox be explained?

icail05.pdf icail05-pres.pdf

The Evolution of Artificial Social Systems

Guido Boella, and Leendert van der Torre

The basic idea of the artificial social systems approach of Shoham and Tennenholtz [1995; 1997] is to add a mechanism, called a social law, that will minimize the need for both centralized control and on-line resolution of conflicts. A social law is defined as a set of restrictions on the agents's activities which allow them enough freedom on the one hand, but at the same time constrain them so that they will not interfere with each other. Several variants have been introduced to reason about the design and emergence of social laws. However, existing models of artificial social systems cannot be used for the evolution of such systems, because these models do not contain an explicit representation of the social laws in force. In this paper we use enforceable social laws [Boella and van der Torre, 2005] to address the question how artificial social systems can be extended to reason about the evolution of artificial social systems.

ijcai05b.pdf

A nonmonotonic logic for specifying and querying preferences

Guido Boella, and Leendert van der Torre

a Preferences are becoming of greater interest in many areas of artificial intelligence, such as knowledge representation, multiagent systems, constraint satisfaction, decision making, and decision-theoretic planning. In the logic of preference there is a debate when a set of preferences should be consistent. For example, Bacchus and Grove [1996] criticize ceteris paribus preferences, because {} should be consistent, and they criticize most existing logics of preference, because {} should be consistent. In order not to restrict the use of the logic of preference, we propose a minimal logic of preference in which any set of specified preferences is consistent. To make it useful for practical applications, we extend this logic to specify preferences with a logic to query preferences, and with a nonmonotonic reasoning mechanism.

ijcai05a.pdf

Roles as a coordination construct: introducing powerJava

Matteo Baldoni, Guido Boella, and Leendert van der Torre

In this paper we apply the role metaphor to coordination. Roles are used in sociology as a way to structure organizations and to coordinate their behavior. In our model, the features of roles are their dependence on an institution, and the powers they assign to players of roles. The institution represents an environment where the components interact with each other by using the powers attributed to them by the roles they play, even when they do not know each other. The interaction between a component playing a role and the role is performed via interfaces stating the requirements to play a role, and which powers are attributed by roles. Roles encapsulate their players' capabilities to interact with the institution and with the other roles, thus achieving separation of concerns between computation and coordination. The institution acts as a coordinator which manages the interactions among components by acting on the roles they play, thus achieving a form of exogenous coordination. As an example, we introduce the role construct in the Java programming language, providing a precompiler for it. In order to better explain the proposal, we show how to use the role construct as a coordination means by applying it to a dining philosophers problem extended with dynamic reconfiguration.

entcs05b.pdf mtcoord05.pdf mtcoord05-pres.ppt mtcoord05-pres.pdf mtcoord05-pres.sxi

Analyzing Control Trust in Normative Multiagent Systems

Joris Hulstijn, Yao-Hua Tan and Leendert van der Torre

It has been argued that transaction trust is composed of party trust and control trust. In this paper we study control trust: trust in an institution that has set up a control mechanism. We present an account of control mechanisms using normative multiagent systems. Control mechanisms consist of constitutive norms which define evidential documents, and regulative norms which define violation conditions and sanctions. The account is illustrated by an analysis of the Letter of Credit trade procedure.

bled05.pdf

Algorithms for a Nonmonotonic Logic of Preferences

Souhila Kaci and Leendert van der Torre

In this paper we introduce and study a nonmonotonic logic to reason about various kinds of preferences. We introduce preference types to choose among these kinds of preferences, based on an agent interpretation. We study ways to calculate "distinguished" preference orders from preferences, and show when these distinguished preference orders are unique. We define algorithms to calculate the distinguished preference orders.

ecsqaru05.pdf

Decision-Theoretic Deliberation in Resource Bounded Self-Aware Agents

Guido Boella, Joris Hulstijn and Leendert van der Torre

In this paper we introduce the notion of decisiontheoretic deliberation, which studies the relation between classical game and decision theory on the one hand, and agent theories of deliberation on the other hand. We aim at modelling the commonsense notion of intention in systems which are self-aware of their bounded reasoning power concerning their decisions. We propose a transparent decision theoretic deliberation model that contains besides standard actions that change the world and standard actions that only change the information state of the agent (capturing the standard notion of the value of information), also actions that change the agenda of the agent. The agenda contains book keeping of the agent's decision theoretic planner, for which we use the DRIPS planner, and is used to deal with bounded resources, such as the trade-off between further deliberation or starting to execute actions, or whether to change the agenda in case of unexpected observations. This model is the basis of our study of intention. Creating an intention corresponds to putting something on the agenda, and reconsidering intentions corresponds to removing things from the agenda. We show that, in contrast to Cohen and Levesque's approach that defines intention in terms of commitment, our treatment of commitment is based in terms of well understood concepts that consider both creation and reconsideration of intention. Consequently, intention cannot be defined locally in a state of the world, but only globally. Moreover, in contrast to abstract approaches such as Rao and Georgeff's characterization of intention in terms of commitment strategies, we relate intention to planning. Finally, in contrast to Bratman's theory of intention, we do not restrict ourselves to classical planning but we incorporate insights from more recent decision-theoretic planning. We illustrate our decision theoretic deliberation model and the related characterization of intention by a detailed example.

CommonSense05.pdf

Enforceable Social Laws

Guido Boella and Leendert van der Torre

In this paper we study the enforcement of social laws in artificial social systems using a control system. We define the enforceable social law problem as an extension of Tennenholtz' stable social law problem. We distinguish the choice of social laws from the choice of control systems, where the latter leads to new computational problems. We consider also properties of sanction based control systems, and monitoring when there is no full observability.

aamas05a.pdf

Reducing Coalition Structures via agreements' representation

Guido Boella, Luigi Sauro and Leendert van der Torre

The concept of power plays an important role in the social sciences and Castelfranchi [4] emphasizes the importance of this concept for multiagent systems. In [2] we build upon this work by distinguishing four viewpoints on multiagent systems: a mind structure, a power structure, a dependence structure and a coalition structure. These viewpoints are increasingly abstract conceptualizations of systems as collections of autonomous cognitive agents. In [1] we propose a way to define coalition structures from power structures. In this paper we refine this approach using task based power views, and we relate it to central notions in game theory.

aamas05b.pdf

Coordination in Normative Multiagent Systems

Guido Boella, Joris Hulstijn and Leendert van der Torre

We survey some of our work in deontic logic and normative multiagent systems. First, we discuss how deontic logic may help the coordination of components by discussing the use of logic as a secretarial assistant, which facilitates exogenous coordination. Secondly, we discuss which concepts normative multiagent systems to offer to coordination, by discussing contracts and roles. Thirdly, we discuss how social commitments help coordination, by discussing the distinction and possible synthesis between the FIPA approach and the social semantics approach.

finco05.pdf

Introduction to Normative Multiagent Systems

Guido Boella, Leendert van der Torre and Harko Verhagen

In this paper we give a short introduction to the emerging area of normative multiagent systems by presenting definitions and examples.

aisb05.pdf

A Game Theoretic Approach to Contracts in Multiagent Systems

Guido Boella and Leendert van der Torre

Contracts are used to create new interaction possibilities among agents, and they therefore play an important role in the game theoretic analysis of agent interaction. We use normative multiagent systems to model both the contracts and the interactions. In particular, we formalize contracts as systems of regulative and constitutive norms within a larger rule-governed setting, and using recursive modelling we develop a game theory where agents make contracts.We show how agents can modify the behavior of normative systems by means of constitutive rules in the contract changing these systems, and we illustrate how agents use the game theory within contract negotiation in organizations.

smcc06.pdf

Using XML transformations for Enterprise Architectures

A. Stam, J. Jacob, F. de Boer, M. Bonsangue, and L. van der Torre,

In this paper we report on the use of XML transformations in the context of Enterprise Architectures. We show that XML transforma- tion techniques are valuable to the analysis, selection and, especially, the visualization of enterprise architectures. We propose a way to describe architectural information within a single XML document. Moreover, we propose transformational techniques to extract views from such a docu- ment and indicate how to perform a specific form of impact analysis on it. The transformations are expressed with the language RML, a compact yet powerful transformation language developed at CWI, which obtains its power from regular expressions defined on XML documents. We dis- cuss a tool that has been built on top of it to visualize the results of the transformations and illustrate the advantages of our approach: the genericity of XML, the application of a single technique (namely XML transformations) for various tasks, and the bene¯ts of having a model viewer which is in complete ignorance of the architectural language used.

isola04.pdf (isola04ws.pdf)

Enterprise Architecture Analysis with XML

Frank de Boer, Marcello Bonsangue, Joost Jacob, Andries Stam, and Leendert van der Torre

This paper shows how XML can be used for static and dynamic analysis of architectures. Our analysis is based on the distinction between symbolic and semantic models of architectures. The core of a symbolic model consists of its signature that specifies symbolically its structural elements and their relationships. A semantic model is defined as a formal interpretation of the symbolic model. This provides a formal approach to the design of architectural description languages and a general mathematical foundation for the use of formal methods in enterprise architectures. For dynamic analysis we define transformations of models of architectures, modeled in XML, and for this purpose the XML vocabulary for an architecture is extended with a few constructs defined in the Rule Markup Language (RML). There are RML tools available that perform the desired transformations.

hicss05.pdf

Virtual Organizations as Normative Multi-agent Systems

Guido Boella, Joris Hulstijn and Leendert van der Torre

In this paper we propose a conceptual model of virtual organizations as normative multiagent systems. The dynamic aspects of virtual organizations are modeled using aspects of speech act theory and Searle's theory of the construction of social reality. We illustrate the use of our model by discussing an example of distributed access control policies. We show how the model captures the distinction between local and global authorities, and between local and global norm enforcement policies.

hicss05b.pdf

Organizations as Socially Constructed Agents in the Agent Oriented Paradigm

Guido Boella and Leendert van der Torre

In this paper we propose a new role for the agent metaphor in the definition of the organizational structure of multiagent systems. The agent metaphor is extended to consider as agents also social entities like organizations, groups and normative systems, so that mental attitudes can be attributed to them - beliefs, desires and goals - and also an autonomous and proactive behavior. We show how the metaphor can be applied also to structure organizations in functional areas and roles, which are described as agents too. Thus, the agent metaphor can play a role similar to the object oriented metaphor which allows structuring objects in component objects. Finally, we discuss how the agent metaphor addresses the problems of control and communication in such structured organizations.

esaw04.pdf

Inferring trust

Mehdi Dastani, Andreas Herzig, Joris Hulstijn, and Leendert van der Torre

In this paper we discuss Liau's logic of Belief, Inform and Trust (BIT), which captures the use of trust to infer beliefs from acquired information. However, the logic does not capture the derivation of trust from other notions. We therefore suggest the following two extensions. First, like Liau we observe that trust in information from an agent depends on the topic of the information. We extend BIT with a formalization of topics which are used to infer trust in a proposition from trust in another proposition, if both propositions have the same topics. Second, for many applications, communication primitives other than inform are required. We extend BIT with questions, and discuss the relationship with belief, inform and trust. An answer to a question can lead to trust, when the answer conforms to the beliefs of the agent.

clima04.pdf

Decision-theoretic deliberation under bounded rationality

Guido Boella, Joris Hulstijn and Leendert van der Torre

Agent theory proposes to model the behavior of complex software systems in terms of mental attitudes like belief, desires, goals, intentions and obligations, ranging from, e.g., the PRS system [7] to the more recent BOID architecture [3]. Decision-theoretic deliberation captures concepts and reasoning mechanisms from agent theory in standard decision-theoretic terms. This is an ambitious enterprise, as it has the objective to bridge the worlds of decision theory and agent theory. Thus far, several partial results on the decision-theoretic characterization have been obtained. The relation between beliefs (as well as defaults) and probabilistic techniques has been studied for some time, there are characterizations of desires and goals in decision-theoretic terms [6], there are various interpretations of obligations and norms, for example as social laws [8], and there are preliminary results on intention [1]. See our comparison paper [4] for an overview.

The most problematic issue in decision-theoretic deliberation is the characterization of intention. Roughly, whereas beliefs have been related to probabilities, desires to utilities, and obligations to social laws, intentions do not seem to have an obvious counterpart in classical game and decision theory. However, most discussions on the popular BDI model have focussed on the role of intention in deliberation [2]. Consequently, we believe that intention is the benchmark example of decision-theoretic deliberation.

loft04.pdf

Attributing Mental Attitudes to Social Entities: Constitutive Rules are Beliefs, Regulative Rules are Goals

Guido Boella and Leendert van der Torre

In this paper, we propose a model of constitutive and regu- lative norms in a logical multiagent framework. We analyze the relation- ship between these two types of rules and explain similarities between them, using the metaphor of considering social entities - like normative systems, groups and organizations - as agents and of attributing them mental attitudes as well as an autonomous behavior.We argue that while constitutive norms expressing "counts-as" relations are modelled as the beliefs of social entities, regulative norms, like obligations, prohibitions and permissions, are modelled as their goals.

collint04.pdf

An Agent Oriented Ontology of Social Reality

Guido Boella and Leendert van der Torre

In this paper we introduce an ontology based on the notion of agent to represent and reason about social reality. We model social constructions as agents, for example, groups, organizations, normative systems, and roles, and we attribute mental attitudes to them. Roughly, we define obligations or regulative norms as goals of the normative system, constitutive norms as beliefs of the normative system, joint, shared, mutual and social beliefs, desires and goal as beliefs, desires and goals of group, responsibilities of an agent as goals of the role he plays, and the required expertise of an agent as beliefs and actions of the role he plays. In this way, we achieve a uniform framework for a large variety of concepts using a small vocabulary, and, in particular, basing it on notions, like mental attitudes, which are commonly used in agent theories. The proposed ontology is modelled using a description logic.

fois04.pdf

Games for Cognitive Agents

Mehdi Dastani and Leendert van der Torre

Strategic games model the interaction among simultaneous decisions of agents. The starting point of strategic games is a set of players (agents) having strategies (decisions) and preferences on the games outcomes. In this paper we do not assume the decisions and preferences of agents to be given in advance, but we derive them from the agents mental attitudes. We specify such agents, define a mapping from their specification to the specification of the strategic game they play. We discuss a reverse mapping from the specification of strategic games that agents play to a specification of those agents. This mapping can be used to specify a group of agents that can play a strategic game, which shows that the notion of agent system specification is expressive enough to play any kind of game.

jelia04.pdf

Local vs Global Policies and Centralized vs Decentralized Control in Virtual Communities of Agents

Guido Boella and Leendert van der Torre

We are interested in the design of policies for virtual communities of agents based on the grid infrastructure. In a virtual community agents can play both the role of resource consumers and the role of resource providers, and they remain in control of their resources. We argue that this requirement creates a distinction between two dimensions: global vs local and centralized and decentralized control by means of policies. The providers should be enabled to specify their local policies on their own resources, but their policies should be consistent with the global policies. At the same time, some aspects of the decentralized control should be delegated to specialized providers; this delegation requires a distinction between the authorization to access a resource and a permission to do so.

wi04.pdf

Fulfilling or Violating Obligations in Normative Multiagent Systems

Guido Boella and Leendert van der Torre

A theory of rational decision making in normative multiagent systems has to distinguish among the many reasons why agents fulfill or violate obligations. We propose a classification of such reasons for single cognitive agent decision making in a single normative system, based on the increasing complexity of this agent. In the first class we only consider the agents motivations, in the second class we consider also its abilities, in the third class we consider also its beliefs, and finally we consider also sensing actions to observe the environment. We sketch how the reasons can be formalized in a normative multiagent system with increasingly complex cognitive agents.

iat04a.pdf

Power and Dependence Relations in Groups of Agents

Guido Boella, Luigi Sauro and Leendert van der Torre

In this paper we present a formal model of Multiagent Systems to analyze the relations of power and dependence underlying group behaviors such as cooperation. Inspired by the work of Castelfranchi we define these relations by means of a description of goals and skills of single agents. We show how our framework can be used to describe social and organizational structures as emergent properties of a collection of individuals.

iat04.pdf

Game Specification in Normative Multiagent System: the Trias Politica

Guido Boella and Leendert van der Torre

In this paper we formalize the specification of games in the trias politica using Rao and Georgeffs specification language BDICTL*. In particular, we generalize Rao and Georgeff's specification of single agent decision trees to multiagent games, for which we introduce observations and recursive modelling, in this setting we formalize obligations, and we characterize four kinds of agents, called legislators, judges, policemen and citizens. Legislators are characterized by their power to create and revise obligations, judges are characterized by their power to count behavior of citizens as violations, and policemen are characterized by their ability to sanction behavior.

iat04.pdf

Persuasion strategies in dialogue

Guido Boella, Joris Hulstijn and Leendert van der Torre

In this abstract we consider dialogues that have a normative aspect. A dialogue is regulated by norms, but can also establish new norms. Certain utterances "count as" a particular dialogue move in some dialogue game, which creates obligations and permissions for the participants. But norms do not operate in isolation; we study their relation to mental attitudes of participants, in particular beliefs (information), desires, goals or intentions.

cmna04.pdf

Power and dependence in multiagent systems

Guido Boella, Luigi Sauro and Leendert van der Torre

An important aim in the field of Multiagent Systems is to study emergent social structures, such as groups and collectives. The relevance of social structures in Distributed Artificial Intelligence, Artificial Life, Sociology necessitates a well motivated definition of their characteristics. The research question of this paper is to formalize the results obtained by Castelfranchi that try to bridge the gap between the BDI model and the macro-level of Multiagent Systems. We presented in (AAMAS04)general social viewpoints on Multiagent Systems and in (ECAI04) we have detailed a derivation of coalition structures from power structures. In this paper we continue the research, and we define power in terms of cognitive (or mind) view, i.e., starting from the characteristics of the single agents as their goals and skills. Properties of collectives, as mutual dependence or cooperation, are also hierarchically defined by means of the definition of power. It is shown that upper levels of this hierarchy need mental model of the agents more and more complex.

focus04.ps

A logical viewpoint on architectures

Frank de Boer, Marcello Bonsangue, Joost Jacob, Andries Stam, and Leendert van der Torre

In this paper we introduce a logical viewpoint on architectures. The logical viewpoint is based on the distinction between symbolic and semantic models of architectures. The core of a symbolic model consists of its signature that specifies symbolically its structural elements and their relationships. A semantic model is defined as a formal interpretation of the symbolic model. This leads to more precise characterization of the concepts introduced in IEEE standard 1471-2000, and provides a formal approach to the design of enterprise of architectural description languages and a general mathematical foundation for the use of formal methods in enterprise architectures.

Additionally, we show how this logical viewpoint allows for the definition of a simple general XML language for the description of both static and dynamic aspects of an architecture. For the meta-analysis of both these aspects we introduce a new XML tool for general XML transformations based on a Rule Markup Language.

edoc04.ps

Concepts for Modelling Enterprise Architectures

Henk Jonkers, Marc Lankhorst, Rene van Buuren, Stijn Hoppenbrouwers, Marcello Bonsangue, and Leendert van der Torre

A coherent description of an enterprise architecture provides insight, enables communication among stakeholders and guides complicated change processes. Unfortunately, so far no enterprise architecture description language exists that fully enables integrated enterprise modelling, because for each architectural domain, architects use their own modelling techniques and concepts, tool support, visualisation techniques, etc. In this paper we outline such an integrated language and we identify and study concepts that relate architectural domains. In our language concepts for describing the relationships between architecture descriptions at the business, application, and technology levels play a central role, related to the ubiquitous problem of businessIT alignment, whereas for each architectural domain we conform to existing languages or standards such as UML. In particular, usage of services offered by one layer to another plays an important role in relating the behaviour aspects of the layers. The structural aspects of the layers are linked through the interface concept, and the information aspects through realisation relations.

ijcis04.pdf

Decisions, deliberation, and agent types: CDT, QDT, BDI, 3APL, BOID

Mehdi Dastani and Leendert van der Torre

In this paper we investigate the relation between decisions, deliberation and agent types. In particular, we are interested how deliberation leads to decisions, and how agent types classify patterns of deliberation. We therefore consider Classical and Qualitative Decision Theories (CDT and QDT), the Beliefs-Desire-Intention (BDI) model, 3APL systems, and Belief-Obligation-Intention-Desire (BOID) systems. The first two are based on a decision rule which expresses a notion of rationality, whereas the latter three are based on deliberation processes and agent types.

focus04.pdf

Attributing Mental Attitudes to Roles: The Agent Metaphor Applied to Organizational Design

Guido Boella and Leendert van der Torre

In this paper we address the problem of defining roles in organizations like e-trade ones. The methodology we use is to model roles according to the agent metaphor: we attribute to roles mental attitudes, like beliefs, desires and goals, we relate them to the agents required expertise and responsibilities, and we model role behavior in game theoretic terms. Analogously, the organization is modelled as an agent which acts as a normative system: it imposes obligations to roles and to the agents playing the roles.

icec04.pdf

Negotiating the distribution of obligations with sanctions among autonomous agents

Guido Boella and Leendert van der Torre

In this paper we study the distribution of obligations together with their associated sanctions among agents belonging to collective entities like groups and organizations. We model the distribution as a negotiation process among the agents, we model the behavior of the agents in a qualitative game theory, and we formalize them in a logical framework. We characterize collective obligations according to the way the responsibility in case of violation is attributed to individual agents or to the whole set of agents, where we distinguish among violations during the negotiation and during the execution of the task. We also show that in some cases it is a drawback to be the only agent able to see to the fulfilment of part of an obligation, but in other cases it may be an advantage, because of the power it gives to the agent during the negotiation.

ecai04.pdf

An Abstraction from Power to Coalition Structures

Guido Boella, Luigi Sauro and Leendert van der Torre

The concept of power plays an important role in the social sciences and Castelfranchi [2, 3] emphasizes the importance of this concept for multiagent systems. In [1] we build upon this work by distinguishing four viewpoints on multiagent systems: a mind structure, a power structure, a dependence structure and a coalition structure. These viewpoints are increasingly abstract conceptualizations of systems as collections of autonomous cognitive agents. In this paper we formally define one kind of power and coalition structures, and an abstraction from this power to its coalition structure.

ecai04b.pdf

Permission and Authorization in Policies for Virtual Communities of Agents

Guido Boella and Leendert van der Torre

We study the design of policies for virtual communities of agents based on peer-to-peer systems or the grid infrastructure. In a virtual community agents can play both the role of resource consumers and the role of resource providers. Moreover, the agents remain in control of their resources, and therefore we distinguish between the authorization to access a resource given by the virtual community and the permission to do so issued by the resource providers. We propose a logical multiagent framework for virtual communities that distinguishes three roles: resource consumption, provision, as well as authorization.

ap2pc04.pdf

Normative multiagent systems and trust dynamics

Guido Boella and Leendert van der Torre

a In this paper we use recursive modelling to formalize sanction-based obligations in a qualitative game theory. In particular, we formalize an agent who attributes mental attitudes such as goals and desires to the normative system which creates and enforces its obligations. The wishes (goals) of the normative system are the commands (obligations) of the agent. Since the agent is able to reason about the normative systems behavior, our model accounts for many ways in which an agent can violate a norm believing that it will not be sanctioned. We thus propose a cognitive theory of normative reasoning which can be applied in theories requiring dynamic trust to understand when it is necessary to revise it.

trust04.pdf (preliminary workshop version)

Combining goal generation and planning in an argumentation framework

Joris Hulstijn and Leendert van der Torre

We study conflicts between goals and plans in Dung's abstract argumentation framework. Argumentation theory has traditionally been used to deal with conflicts between defaults and beliefs. Recently Amgoud has proposed to use it for conflicts between plans. Amgoud argues that Dung's argumentation theory has to be adjusted, because conflicts between plans are fundamentally different from conflicts between defaults. We agree with the fundamental difference, but we propose an alternative way to deal with conflicts between plans that stays within Dung's framework. Moreover, we extend Amgoud s argumentation framework for planning with goal generation procedures, that can deal with conflicts between goals. In the proposed framework goals are derived from desires by forward reasoning, and plans are derived from goals and planning rules by backward reasoning.

nmr04.pdf

Groups as Agents with Mental Attitudes

Guido Boella and Leendert van der Torre

We discuss a model of cooperation among autonomous agents, based on the attribution of mental attitudes to groups: these attitudes represent the shared beliefs and objectives and the wish to reduce the costs for the members. When agents take a decision they have to recursively model what their partners are expected to do under the assumption that they are cooperative, and they have to adopt the goals and desires attributed to the group: otherwise, the other members consider them uncooperative and thus liable.

aamas04a.pdf aamas04a.ps

Contracts as Legal Institutions in Organizations of Autonomous Agents

Guido Boella and Leendert van der Torre

In this paper we address the problem of how the autonomy of agents in an organization can be enhanced by means of contracts. Contracts are modelled as legal institutions: systems of legal rules which allow to change the regulative and constitutive norms of an organization. The methodology we use is to attribute to organizations mental attitudes, beliefs, desires and goals, and to take into account their behavior by using recursive modelling.

aamas04b.pdf aamas04b.ps

Social Viewpoints on Multiagent Systems

Guido Boella, Luigi Sauro, and Leendert van der Torre

An important aim in the field of multiagent systems is to study emergent social structures, such as sets and collectives. Castelfranchi introduces concepts like groups and collectives from social theory in agent theory, both to enrich agent theory and to develop experimental, conceptual and theoretical new instruments for the social sciences. In this paper we contribute to this project by addressing the following questions: