Dissertations / Theses on the topic 'Open Agent System'

COORDENAÇÃO DE APERFEIÇOAMENTO DO PESSOAL DE ENSINO SUPERIOR<br>Nesta dissertação apresenta-se um modelo para aplicação de contratos em ambientes multi-agentes abertos. A idéia principal em um sistema multiagente é que um comportamento global inteligente possa ser alcançado a partir do comportamento individual dos agentes. Neste contexto, surge a dificuldade em se garantir que agentes estão cooperando corretamente para alcançar os objetivos da organização no qual estão inseridos. O modelo proposto neste trabalho visa expandir os conceitos de contratos em componentes a sistemas multi-agentes abertos. Contratos podem ser entendidos como uma forma de negociação entre componentes, no qual se estabelecem obrigações e benefícios que devem ser respeitados por todos os participantes do contrato na execução de serviços. Porém, quando o assunto é sistemas multi-agentes, há pouca literatura disponível. Isto se deve pela dificuldade de se traduzir contratos para componentes sobre o paradigma orientado a objetos às características dos agentes. Componentes possuem métodos e interfaces bem definidas, enquanto agentes ocultam suas estruturas internas e talvez apresentem comportamentos complexos. Um framework foi desenvolvido baseado no modelo conceitual proposto. O resultado visa demonstrar a viabilidade de se aplicar contratos para componentes a ambientes multi-agentes abertos, com o objetivo de gerenciar e averiguar a cooperação entre agentes, levando em consideração seus papéis na organização e respeitando suas características individuais.<br>In this work, we propose a model for applying contracts in open multi-agent systems. The main idea in a multi-agent system is that an intelligent global behavior can be reached from the individual behavior of the agents. In this context, it is difficulty to guarantee that the agents are correctly cooperating to reach the organization objectives in which they are inserted. The model considered in this work expands the contract concepts in components to open multi-agent system. Contracts can be understood as a negotiation form between components, which entails obligations and benefits for both parties. However, when the subject is multi-agent systems, there are a few available literatures. This is explained because contracts for components, guided on object paradigm, are hard to be translated to the characteristics of the agents. Components have methods and well defined interfaces, while agents hide their internal structures and perhaps they present complex behaviors. A framework was developed based on the conceptual model we are proposing. The result demonstrates the viability of applying contracts for components to the open multi-agents environments. The main goal is to manage and to inquire the cooperation between agents, considering the agents roles in the organization and respecting the individual agent characteristics.

As software requirements grow increasingly complex, the need to connect to and re-use existing, tested software, grows with it. Open systems, such as the Internet, aid this process by connecting together software services provided by a range of organisations, and the distributed nature of the system allows the services to be regularly updated and improved. Applications can be deployed within the open systems that opportunistically attempt to make use of the best functionality available at any one time. Agent-based systems have been proposed as an ideal way to implement such applications, due to their flexibility and distributed control. However, a balance must be kept between acting opportunistically and ensuring that each application operates to the standards demanded by the application requirements. Determining whether an application will perform to its requirements necessitates justifying the design decisions made in creating it. Our goal is to provide application developers with the means to create justified designs for opportunistic applications. The main contribution of this thesis is a software engineering methodology, agent interaction analysis, based on a set of independently valuable techniques we have developed. The first of these is a novel approach to modelling applications as being instantiated by a set of agent interactions, allowing such applications to be described with minimal restrictions on their implemented structure. Second, we provide a technique, based on design patterns, for comparing mechanisms for instantiating parts of multi-agent system. Finally, we provide an approach to more detailed analysis and comparison of coordination mechanisms.

During the last decade, the problem of consensus in Multi-Agent Systems (MASs) has been studied with special emphasis on graph theoretical methods. Consensus can be regarded as a control objective in which all the agents in a network converge to (or agree “upon”) a common value. This is achieved through a given control strategy usually referred to as consensus algorithm. The consensus problem in MASs becomes more challenging in the presence of agents that can join and leave the network, hence the so-called “open” multi-agent systems. This topic is recent in the scientific literature and quickly gaining attention. Thus, this dissertation is motivated by this topic, focusing on the consensus problem on the median consensus value and the average consensus in open multi-agent systems. Hence, this thesis is first devoted to designing a novel distributed open average consensus protocol for multi-agent systems. The distributed algorithm tracks the average of the agents’ state despite the time-varying size and composition of the network. The research activity consists of the design and formal characterization of the convergence properties of the algorithm. The results have been corroborated by numerical simulations. As for the second part of the research activity, the characterization of the convergence properties of a distributed algorithm to compute and track the median value of a set of numbers in open multi-agent systems has been proposed. A continuous time formulation is considered where the state variables of the agents track with zero error the median value of a set of time-varying reference signals given as input to the agents in a time-varying, undirected network topology. The performance of the proposed protocol is considered in the framework of open multi-agent systems by proposing join and leave mechanisms, i.e., the scenario where agents may join and leave the network during the protocol execution. One notable feature of consensus on the median value is the robustness of the median value, as opposed to the average value, with respect to abnormal or outlier state values. Non-smooth Lyapunov theory is employed to provide convergence guarantees and simple tuning rules to adjust the algorithm parameters. Apart from studying open multi-agent systems, this dissertation also proposes a distributed scheme for transforming any connected interaction graph with a possible no integer average degree into a connected approximately random k-regular graph which is independent of the degree of the initial graph. in more details, we define some local graph transformation rules (consisting of rules for cutting, adding, and moving edges) and provide a distributed implementation. In the resulting process, a random regular graph is obtained while the agents observe and modify only the local structure of the network. As such, the network achieves high expansion ratios and algebraic connectivity, which provide robustness to various structural and functional perturbations.

CONSELHO NACIONAL DE DESENVOLVIMENTO CIENTÍFICO E TECNOLÓGICO<br>Sistemas multi-agentes abertos são freqüentemente caracterizados como sistemas distribuídos onde agentes, que podem não ser conhecidos a priori, podem entrar ou sair do sistema a qualquer momento. Uma abordagem de governança estabelece regras de interação que precisam ser seguidas pelos agentes de um determinado sistema. O uso destas regras permite uma maior previsibilidade e controle sobre o comportamento observável do sistema. Nesta tese, apresenta-se uma abordagem de governança de sistemas multi-agentes abertos como adequada para lidar não apenas com o monitoramento e controle do comportamento dos agentes, mas também com aspectos de fidedignidade. Um software é dito fidedigno quando se pode confiar no mesmo através de verificações formais ou informais assumindo riscos de danos compatíveis com o serviço prestado pelo software. A incorporação de aspectos de fidedignidade em uma abordagem de governança tem como principal benefício a geração de uma tecnologia integrada que possui as vantagens de uma abordagem de governança e ao mesmo tempo lida com especificações de instrumentos de fidedignidade tais como prevenção e tolerância a faltas. A abordagem proposta é ilustrada através de um estudo de caso no contexto de controle de tráfego aéreo.<br>Open multi-agent systems are frequently characterized by having little or no control over the behavior of the agents. The internal implementation and architecture of agents usually are inaccessible, and different teams may have developed them but with no coordination between them. Furthermore, agents may enter or leave the system at their will. A governance approach defines the interaction rules that must be obeyed by the agents. These rules allow for a greater control and predictability of the observable system behavior. In this thesis, we propose a governance approach to deal not only with the monitoring and control of agents` behavior but also to deal with dependability concerns. The original definition of dependability is the ability to deliver service that can justifiably be trusted. A governance approach that also deals with dependability has as main benefit the reuse of the monitoring and enforcement present in the governance infrastructure for dependability. We present a case study in the context of an air traffic control system to illustrate our approach.

Trust and reputation are central to effective interactions in open multi-agent systems (MAS) in which agents, that are owned by a variety of stakeholders, continuously enter and leave the system. This openness means existing trust and reputation models cannot readily be used since their performance suffers when there are various (unforseen) changes in the environment. To this end, this thesis develops and evaluates FIRE, a trust and reputation model that enables autonomous agents in open MAS to evaluate the trustworthiness of their peers and to select good partners for interactions. FIRE integrates four sources of trust information under the same framework in order to provide a comprehensive assessment of an agent’s likely performance in open systems. Specifically, FIRE incorporates interaction trust, role-based trust, witness reputation, and certified reputation, that models trust resulting from direct experiences, role-based relationships, witness reports, and third-party references, respectively, to provide trust metrics in most circumstances. A novel model of reporter credibility has also been integrated to enable FIRE to effectively deal with inaccurate reports (from witnesses and referees). Finally, adaptive techniques have been introduced, which make use of the information gained from monitoring the environment, to dynamically adjust a number of FIRE’s parameters according to the actual situation an agent finds itself in. In all cases, a systematic empirical analysis is undertaken to evaluate the effectiveness of FIRE in terms of the agent’s performance.

Internet es, tal vez, el avance científico más relevante de nuestros días. Entre otras cosas, Internet ha permitido la evolución de los paradigmas de computación tradicionales hacia el paradigma de computaciónn distribuida, que se caracteriza por utilizar una red abierta de ordenadores. Los sistemas multiagente (SMA) son una tecnolog a adecuada para abordar los retos motivados por estos sistemas abiertos distribuidos. Los SMA son aplicaciones formadas por agentes heterog eneos y aut onomos que pueden haber sido dise~nados de forma independiente de acuerdo con objetivos y motivaciones diferentes. Por lo tanto, no es posible realizar ninguna hip otesis a priori sobre el comportamiento de los agentes. Por este motivo, los SMA necesitan de mecanismos de coordinaci on y cooperaci on, como las normas, para garantizar el orden social y evitar la aparici on de conictos. El t ermino norma cubre dos dimensiones diferentes: i) las normas como un instrumento que gu a a los ciudadanos a la hora de realizar acciones y actividades, por lo que las normas de nen los procedimientos y/o los protocolos que se deben seguir en una situaci on concreta, y ii) las normas como ordenes o prohibiciones respaldadas por un sistema de sanciones, por lo que las normas son medios para prevenir o castigar ciertas acciones. En el area de los SMA, las normas se vienen utilizando como una especi caci on formal de lo que est a permitido, obligado y prohibido dentro de una sociedad. De este modo, las normas permiten regular la vida de los agentes software y las interacciones entre ellos. La motivaci on principal de esta tesis es permitir a los dise~nadores de los SMA utilizar normas como un mecanismo para controlar y coordinar SMA abiertos. Nuestro objetivo es elaborar mecanismos normativos a dos niveles: a nivel de agente y a nivel de infraestructura. Por lo tanto, en esta tesis se aborda primero el problema de la de nici on de agentes normativos aut onomos que sean capaces de deliberar acerca<br>Criado Pacheco, N. (2012). Using Norms To Control Open Multi-Agent Systems [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/17800<br>Palancia

Infrastructure as a Service (IaaS) is a cloud service provisioning model which largely focuses on data centre provisioning of computing and storage facilities. The networking aspects of IaaS beyond the data centre are a limiting factor preventing communication services that are sensitive to network characteristics from adopting this approach. Cloud networking is a new technology which integrates network provisioning with the existing cloud service provisioning models thereby completing the cloud computing picture by addressing the networking aspects. In cloud networking, shared network resources are virtualized, and provisioned to customers and end-users on-demand in an elastic fashion. This technology allows various kinds of optimization, e.g., reducing latency and network load. Further, this allows service providers to provision network performance guarantees as a part of their service offering. However, this new approach introduces new security challenges. Many of these security challenges are addressed in the CloNe security architecture. This thesis presents a set of potential techniques for securing different resource in a cloud network environment which are not addressed in the existing CloNe security architecture. The thesis begins with a holistic view of the Cloud networking, as described in the Scalable and Adaptive Internet Solutions (SAIL) project, along with its proposed architecture and security goals. This is followed by an overview of the problems that need to be solved and some of the different methods that can be applied to solve parts of the overall problem, specifically a comprehensive, tightly integrated, and multi-level security architecture, a key management algorithm to support the access control mechanism, and an intrusion detection mechanism. For each method or set of methods, the respective state of the art is presented. Additionally, experiments to understand the performance of these mechanisms are evaluated on a simple cloud network test bed. The proposed key management scheme uses a hierarchical key management approach that provides fast and secure key update when member join and member leave operations are carried out. Experiments show that the proposed key management scheme enhances the security and increases the availability and integrity. A newly proposed genetic algorithm based feature selection technique has been employed for effective feature selection. Fuzzy SVM has been used on the data set for effective classification. Experiments have shown that the proposed genetic based feature selection algorithm reduces the number of features and hence decreases the classification time, while improving detection accuracy of the fuzzy SVM classifier by minimizing the conflicting rules that may confuse the classifier. The main advantages of this intrusion detection system are the reduction in false positives and increased security.<br>Infrastructure as a Service (IaaS) är en Cloudtjänstmodell som huvudsakligen är inriktat på att tillhandahålla ett datacenter för behandling och lagring av data. Nätverksaspekterna av en cloudbaserad infrastruktur som en tjänst utanför datacentret utgör en begränsande faktor som förhindrar känsliga kommunikationstjänster från att anamma denna teknik. Cloudnätverk är en ny teknik som integrerar nätverkstillgång med befintliga cloudtjänstmodeller och därmed fullbordar föreställningen av cloud data genom att ta itu med nätverkaspekten.  I cloudnätverk virtualiseras delade nätverksresurser, de avsätts till kunder och slutanvändare vid efterfrågan på ett flexibelt sätt. Denna teknik tillåter olika typer av möjligheter, t.ex. att minska latens och belastningen på nätet. Vidare ger detta tjänsteleverantörer ett sätt att tillhandahålla garantier för nätverksprestandan som en del av deras tjänsteutbud. Men denna nya strategi introducerar nya säkerhetsutmaningar, exempelvis VM migration genom offentligt nätverk. Många av dessa säkerhetsutmaningar behandlas i CloNe’s Security Architecture. Denna rapport presenterar en rad av potentiella tekniker för att säkra olika resurser i en cloudbaserad nätverksmiljö som inte behandlas i den redan existerande CloNe Security Architecture. Rapporten inleds med en helhetssyn på cloudbaserad nätverk som beskrivs i Scalable and Adaptive Internet Solutions (SAIL)-projektet, tillsammans med dess föreslagna arkitektur och säkerhetsmål. Detta följs av en översikt över de problem som måste lösas och några av de olika metoder som kan tillämpas för att lösa delar av det övergripande problemet. Speciellt behandlas en omfattande och tätt integrerad multi-säkerhetsarkitektur, en nyckelhanteringsalgoritm som stödjer mekanismens åtkomstkontroll och en mekanism för intrångsdetektering. För varje metod eller för varje uppsättning av metoder, presenteras ståndpunkten för respektive teknik. Dessutom har experimenten för att förstå prestandan av dessa mekanismer utvärderats på testbädd av ett enkelt cloudnätverk. Den föreslagna nyckelhantering system använder en hierarkisk nyckelhantering strategi som ger snabb och säker viktig uppdatering när medlemmar ansluta sig till och medlemmarna lämnar utförs. Försöksresultat visar att den föreslagna nyckelhantering system ökar säkerheten och ökar tillgänglighet och integritet. En nyligen föreslagna genetisk algoritm baserad funktion valet teknik har använts för effektiv funktion val. Fuzzy SVM har använts på de uppgifter som för effektiv klassificering. Försök har visat att den föreslagna genetiska baserad funktion selekteringsalgoritmen minskar antalet funktioner och därmed minskar klassificering tiden, och samtidigt förbättra upptäckt noggrannhet fuzzy SVM klassificeraren genom att minimera de motstående regler som kan förvirra klassificeraren. De främsta fördelarna med detta intrångsdetekteringssystem är den minskning av falska positiva och ökad säkerhet.

One way to build large-scale autonomous systems is to develop an open multi-agent system using peer-to-peer architectures in which agents are not pre-engineered to work together and in which agents themselves determine the social norms that govern collective behaviour. The social norms and the agent interaction models can be described by Electronic Institutions such as those expressed in the Lightweight Coordination Calculus (LCC), a compact executable specification language based on logic programming and pi-calculus. Open multi-agent systems have experienced growing popularity in the multi-agent community and are expected to have many applications in the near future as large scale distributed systems become more widespread, e.g. in emergency response, electronic commerce and cloud computing. A major practical limitation to such systems is security, because the very openness of such systems opens the doors to adversaries for exploit existing vulnerabilities. This thesis addresses the security of open multi-agent systems governed by electronic institutions. First, the main forms of attack on open multi-agent systems are introduced and classified in the proposed attack taxonomy. Then, various security techniques from the literature are surveyed and analysed. These techniques are categorised as either prevention or detection approaches. Appropriate countermeasures to each class of attack are also suggested. A fundamental limitation of conventional security mechanisms (e.g. access control and encryption) is the inability to prevent information from being propagated. Focusing on information leakage in choreography systems using LCC, we then suggest two frameworks to detect insecure information flows: conceptual modeling of interaction models and language-based information flow analysis. A novel security-typed LCC language is proposed to address the latter approach. Both static (design-time) and dynamic (run-time) security type checking are employed to guarantee no information leakage can occur in annotated LCC interaction models. The proposed security type system is then formally evaluated by proving its properties. A limitation of both conceptual modeling and language-based frameworks is difficulty of formalising realistic policies using annotations. Finally, the proposed security-typed LCC is applied to a cloud computing configuration case study, in which virtual machine migration is managed. The secrecy of LCC interaction models for virtual machine management is analysed and information leaks are discussed.

Dans cette thèse, nous abordons différents problèmes formulés dans un scénario de système multi-agents ouvert (OMAS) où l'ensemble des agents peut changer dans le temps, indépendamment de l'évolution de la dynamique associée au système. Nous considérons des OMAS formulés à l'aide d'un réseau fixe de taille finie, et utilisons deux approches distinctes pour les analyser. Dans la première approche, nous considérons des scénarios caractérisés par l'activation et la désactivation d'agents, de sorte qu'à chaque instant un sous-ensemble différent d'agents actifs peut interagir dans le système. Dans la deuxième approche, nous étudions des scénarios caractérisés par des remplacements d'agents où, à un instant donné, un agent peut être remplacé alors que les autres ne changent pas. Dans ce cas, tous les agents peuvent interagir à tout moment.Trois problèmes différents sont considérés dans cette thèse : le consensus randomisé, le problème d'allocation des ressources et les épidémies. Premièrement, nous analysons le problème du consensus randomisé soumis à un bruit additif où différents sous-ensembles d'agents échangent des informations à chaque itération. Nous définissons un indice de bruit basé sur l'erreur quadratique moyenne attendue et nous dérivons des bornes supérieures. Ensuite, nous considérons le problème d'allocation de ressources où les agents peuvent être remplacés lors de l'implémentation d'un algorithme d'optimisation. Pour ce problème d'optimisation, nous analysons deux algorithmes différents : la descente de gradient pondérée (weighted gradient descent) et la descente de coordonnées aléatoires (random coordinate descent). Pour la descente de gradient pondérée, nous évaluons les performances de l'algorithme dans un OMAS soumis à des pertes de paquets en définissant des métriques de performances appropriées. Pour l'algorithme de descente de coordonnées aléatoires, nous étudions la convergence vers le minimiseur dans un OMAS et nous proposons une analyse alternative à l'aide d'outils inspirés de l'optimisation en ligne. Enfin, nous étudions une épidémie SIS en temps continu sujette à des remplacements d'agents lors de sa propagation. Nous effectuons l'analyse en utilisant une fonction d'agrégation et en dérivant des bornes supérieures pour son comportement asymptotique<br>In this thesis we address several problems formulated in an open multi-agent system (OMAS) scenario where the set of agents can change in time, independently of the evolution of the dynamics associated with the system. We analyze the case of OMAS formulated in a fixed finite size network and we use two approaches for the analysis of the systems. In the first approach we consider scenarios characterized by activation/deactivation of agents such that at each time instant a different subset of active agents can interact in the system. In the second approach we study scenarios characterized by replacements of agents where at a specific time instant, an agent can be replaced while the rest of the agents remain the same. In this case, all the agents are able to interact at all time.Three different problems are considered in this thesis: randomized consensus, resource allocation problem and epidemics. First, we analyze the problem of randomized consensus subject to additive noise where different subset of agents exchange information at each iteration. We define a noise index based on the expected mean squared error and we derive upper bounds. Then, we consider the resource allocation problem where agents can be replaced during the implementation of an optimization algorithm. For this problem, we analyze two different algorithms: weighted gradient descent and random coordinate descent. For the weighted gradient descent, we evaluate the performance of the algorithm in an OMAS subject to packet losses by defining appropriate performance metrics. For the random coordinate descent algorithm, we study the convergence to the minimizer in an OMAS and we provide an alternative analysis using tools inspired from online optimization. Finally, we study a SIS epidemic in continuous time subject to replacements of agents during the evolution of the disease. We perform the analysis using an aggregate function and deriving upper bounds for its asymptotic behavior

Multi Agent Systems (MAS) are computerised systems composed of autonomous software agents that interact to solve complex problems. Within a MAS, agents require some mechanism to coordinate their activities. In the MAS literature, norms have been widely used to coordinate agents’ activities. Thus, given a MAS, a major research challenge is how to synthesise a normative system, namely a collection of norms, which supports its agents’ coordination. This dissertation focuses on the automated synthesis of norms for open Multi- Agent Systems. In an open MAS, the agent population may change along time, agents may be developed by third parties and their behaviours are not known beforehand. These particular conditions make specially challenging to synthesise a normative system to govern an open MAS. The MAS literature has mainly investigated two general approaches to norm synthesis: off-line design, and on-line synthesis. The first approach aims at synthesising a normative system at design time. With this aim, it assumes that the MAS state space is known at design time and does not change at runtime. This goes against the nature of open MAS, and thus off-line design is not appropriate to synthesise their norms. Alternatively, on-line norm synthesis considers that norms are synthesised at runtime. Most on-line synthesis research has focused on norm emergence, which considers that agents synthesise their own norms, thus assuming that they have norm synthesis capabilities. Again, this cannot be assumed in open MAS. Against this background, this dissertation introduces a whole computational framework to perform on-line norm synthesis for open Multi-Agent Systems. Firstly, this framework provides a computational model to synthesise norms for a MAS at runtime. Such computational model requires neither knowledge about agents’ behaviours beforehand nor their participation in the norm synthesis pro- cess. Instead, it considers a regulatory entity that observes agents’ interactions at runtime, identifying situations that are undesirable for coordination to sub- sequently synthesise norms that regulate these situations. Our computational model has been conceived to be of general purpose so that it can be employed to synthesise norms in a wide range of application domains by providing little domain-dependent information. Secondly, our framework provides an abstract architecture to implement such regulatory entity (the so-called Norm Synthesis Machine), which observes a MAS and executes a synthesis strategy to synthe- sise norms. Thirdly, our framework encompasses a family of norm synthesis strategies intended to be executed by the Norm Synthesis Machine. Overall, this family of strategies supports multi-objective on-line norm synthesis Our first synthesis strategy, the so-called base, aims at synthesising effective normative systems that successfully avoid situations that are undesirable for a MAS’ coordination. Then, two further strategies (called iron and simon) go beyond effectiveness and also consider compactness as a norm synthesis goal. iron and simon take alternative approaches to synthesise compact normative systems that, in addition to effectively achieve coordination, are as synthetic as possible. This allows them to reduce agents’ computational efforts when reasoning about norms. A fourth strategy, the so-called lion, goes beyond effectiveness and compactness to also consider liberality as a synthesis goal. lion aims at synthesising normative systems that are effective and compact while preserving agents’ freedom to the greatest possible extent. Our final strategy is desmon, which is capable of synthesising norms by considering different degrees of reactivity. desmon allows to adjust the amount of information that is required to decide whether a norm must be included in a normative system or not. Thus, desmon can synthesise norms either by being reactive (i.e., by considering little information), or by being more deliberative (by employing more information). We provide empirical evaluations of our norm synthesis strategies in two application domains: a road traffic domain, and an on-line community domain. In this former domain, we employ these strategies to synthesise effective, compact and liberal normative systems that successfully avoid collisions between cars. In the latter domain, our strategies synthesise normative systems based on users’ complaints about inappropriate contents. In this way, our strategies implement a regulatory approach that synthesises norms when there is enough user consensus about the need for norms. Overall, this thesis advances in the state of the art in norm synthesis by providing a novel computational model, an abstract architecture and a family of strategies for on-line norm synthesis for open Multi-Agent Systems.<br>Els sistemes Multi-Agent (MAS) són sistemes computeritzats composats d’agents autònoms que interaccionen per resoldre problemes complexos. A un MAS, els agents requereixen algun mecanisme per a coordinar les seves activitats. A la literatura en Sistemes Multi-Agent, les normes han estat àmpliament utilitzades per coordinar les activitats dels agents. Per tant, donat un MAS, un dels majors reptes d’investigació és el de sintetizar el sistema normatiu, és a dir, la col·lecció de normes, que suporti la coordinació dels agents. Aquesta tesi es centra en la síntesi automàtica de normes per sistemes Multi-Agent oberts. A un MAS obert, la població d’agents pot canviar amb el temps, els agents poden ésser desenvolupats per terceres parts, i els comportaments dels agents són desconeguts per endavant. Aquestes condicions particulars fan especialment complicat sintetizar el sistema normatiu que reguli un sistema Multi-Agent obert. En general, la literatura en Sistemes Multi-Agent ha investigat dues aproximacions a la síntesi de normes: disseny off-line, i síntesi on-line. La primera aproximació consisteix a sintetizar un sistema normatiu en temps de disseny. Amb aquest propòsit, aquesta aproximació assumeix que l’espai d’estats d’un MAS és conegut en temps de disseny i no canvia en temps d’execució. Això va contra la natura dels sistemes Multi-Agent oberts, i per tant el disseny off-line no és apropiat per a sintetitzar les seves normes. Com a alternativa, la síntesi on-line considera que les normes són sintetizades en temps d’execució. La majoria de recerca en síntesi on-line s’ha centrat en la emergència de normes, que considera que els agents sintetizen les seves pròpies normes, per tant assumint que tenen la capacitat de sintetitzar-les. Aquestes condicions tampoc no es poden assumir en un MAS obert. Donat això, aquesta tesi introdueix un marc computacional per la síntesi on-line de normes en sistemes Multi-Agent oberts. Primer, aquest marc proveeix un model computacional per sintetizar normes per un MAS en temps d’execució. Aquest model computacional no requereix ni coneixement sobre els comportaments dels agents per endavant ni la seva participación en la síntesi de normes. En canvi, considera que una entitat reguladora observa les interaccions dels agents en temps d’execució, identificant situacions indesitjades per la coordinació i sintetizant normes que regulen aquestes situacions. El nostre model computacional ha estat dissenyat per a ésser de propòsit general per tal que pugui ser utilitzat a la síntesi de normes en un ampli ventall de dominis d’aplicació proporcionant només información clau sobre el domini. Segon, el nostre marc proveeix una arquitectura abstracta per implementar aquesta entitat reguladora, anomenada Màquina de Síntesi, que observa un MAS en temps d’execució i executa una estratègia de síntesi que s’encarrega de sintetizar normes. Tercer, el nostre marc incorpora una familia d’estratègies de síntesi destinades a ésser executades per una màquina de síntesi. En general, aquesta familia d’estratègies soporta la síntesi multi-objectiu i on-line de normes. La nostra primera estratègia, anomenada BASE, està dissenyada per sintetitzar sistemes normatius eficaços que evitin de manera satisfactòria situacions indesitjades per la coordinació d’un sistema Multi-Agent. Després, dues estratègies de síntesi, anomenades IRON i SIMON, van més enllà de la eficàcia i també consideren la compacitat com a objectiu de síntesi. IRON i SIMON prenen aproximacions alternatives a la síntesi de sistemes normatius compactes que, a més d’aconseguir la coordinació de manera efectiva, siguin tant sintètics com fos possible. Això permet a aquestes estratègies reduir els esforços computacionals dels agents a l’hora de raonar sobre les normes. Una quarta estratègia, anomenada LION, va més enllà de la eficàcia i la compacitat per considerar també la liberalitat com a objectiu de síntesi. Lion sintetitza sistemes normatius que són eficaços i compactes mentre preserven la llibertat dels agents tant com sigui possible. La nostra última estratègia és desmon, que és capaç de sintetizar normes considerant diferents graus de reactivitat. desmon permet ajustar la quantitat d’informació necessària per decidir si una norma cal que sigui o no inclosa a un sistema normatiu. DESMON pot sintetizar normes essent reactiu (considerant poca informació), o essent més deliberatiu (considerant més informació). En aquesta tesi presentem avaluacions empíriques de les nostres estratègies de síntesi en dos dominis d’aplicació: el domini del tràfic, i el domini de les comunitats on-line. En aquest primer domini, utilitzem les nostres estratègies per a sintetizar sistemes normatius eficaços, compactes i liberals que eviten colisions entre cotxes. Al segon domini, les nostres estratègies sintetizen sistemes normatius basant-se en les queixes dels usuaris de la comunitat sobre continguts inapropiats. D’aquesta manera, les nostres estratègies implementen un mecanisme de regulació que sintetiza normes quan hi ha suficient consens entre els usuaris sobre la necessitat de normes. Aquesta tesi avança en l’estat de l’art en síntesi de normes al proporcionar un novedós model computacional, una arquitectura abstracta i una familia d’estratègies per la síntesi on-line de normes per sistemes Multi-Agent oberts.

Cooperation and coordination between agents are fundamental processes for increasing aggregate and individual benefit in open Multi-Agent Systems (MAS). The increased ubiquity, size, and complexity of open MAS in the modern world has prompted significant research interest in the mechanisms that underlie cooperative and coordinated behaviour. In open MAS, in which agents join and leave freely, we can assume the following properties: (i) there are no centralised authorities, (ii) agent authority is uniform, (iii) agents may be heterogeneously owned and designed, and may consequently have con icting intentions and inconsistent capabilities, and (iv) agents are constrained in interactions by a complex connecting network topology. Developing mechanisms to support cooperative and coordinated behaviour that remain effective under these assumptions remains an open research problem. Two of the major mechanisms by which cooperative and coordinated behaviour can be achieved are (i) trust and reputation, and (ii) norms and conventions. Trust and reputation, which support cooperative and coordinated behaviour through notions of reciprocity, are effective in protecting agents from malicious or selfish individuals, but their capabilities can be affected by a lack of information about potential partners and the impact of the underlying network structure. Regarding conventions and norms, there are still a wide variety of open research problems, including: (i) manipulating which convention or norm a population adopts, (ii) how to exploit knowledge of the underlying network structure to improve mechanism efficacy, and (iii) how conventions might be manipulated in the middle and latter stages of their lifecycle, when they have become established and stable. In this thesis, we address these issues and propose a number of techniques and theoretical advancements that help ensure the robustness and efficiency of these mechanisms in the context of open MAS, and demonstrate new techniques for manipulating convention emergence in large, distributed populations. Specfically, we (i) show that gossiping of reputation information can mitigate the detrimental effects of incomplete information on trust and reputation and reduce the impact of network structure, (ii) propose a new model of conventions that accounts for limitations in existing theories, (iii) show how to manipulate convention emergence using small groups of agents inserted by interested parties, (iv) demonstrate how to learn which locations in a network have the greatest capacity to in uence which convention a population adopts, and (v) show how conventions can be manipulated in the middle and latter stages of the convention lifecycle.

Large-scale networked environments, such as the Internet, possess the characteristics of distributed data, distributed access and distributed control; this gives the user a powerful mechanism for building and integrating large repositories of distributed information from diverse resource sets. However, few support tools have been developed to allow the user to take advantage of the distributed nature of their information. Distributed information management is the process by which users can create, disseminate, discover and manage information that is spread across distributed resources. Distributed open hypermedia systems have shown how distributed information, such as documents and hypermedia links, can be managed and handled within an environment that integrates smoothly between the user's desktop and the network. However, such systems are now looking at addressing the problem of interoperability across hypermedia systems, so that documents and links can be shared between users on heterogeneous integrating technologies. This thesis proposes that the distributed information management provided by open hypermedia systems needs to be extended so that it is more interoperable, extensible and pervasive and that this can be achieved by integrating the principles of open hypermedia with the technology of mobile agents. Mobile agents present a new development mechanism for designing and building distributed applications which are well suited to the dynamic environment of large-scale networks. This thesis describes the development of a mobile agent architecture within which distributed information management tasks can be built and executed. Mobile agents present an important abstraction mechanism when designing distributed environments and also allow the user to manage distributed information indirectly through their mobile agents. A number of prototype agents are described that have been developed to illustrate distributed information management tasks within the architecture and to show how abstractionism and indirect management can be achieved.

Dans cette thèse, nous essayons de fournir des outils automatisés pour élaborer des systèmes ouverts sûrs. Les systèmes ouverts, qui peuvent être vus comme des systèmes multi-agents, peuvent être spécifiés en ATL. La logique ATL a été introduite dans ce but précis. Il existe deux extensions intéressantes d'ATL, à savoir ATL+ et ATL* (ATL+ étant une restriction d'ATL*). ATL+ permet la combinaison Booléenne d'opérateurs temporels, et ATL* permet également l'imbrication d'opérateurs temporels. La procédure de décision basée sur les tableaux pour ATL est une méthode constructive pour tester la satisfiabilité d'une spécification donnée. Elle est constructive dans le sens qu'il est possible d'extraire un modèle depuis le tableau obtenu, lorsque la formule de départ est satisfiable. Dans cette thèse, nous proposons deux procédures de décision basées sur les tableaux pour ATL+ and ATL*, ainsi qu'une implémentation de ces procédures. Notre procédure est correcte, complète et optimale. En effet, nos deux procédures s'exécutent en 2Exptime. A notre connaissance, notre implémentation est le premier exécutable pour décider la satisfiabilité des formules ATL et ATL*. En perspective de cette thèse, nous discutons de la possibilité d'améliorer l'extraction de modèles depuis les tableaux pour ATL, ATL+ and ATL*. Nous aimerions obtenir à la fin des modèles relativement petits<br>In this thesis, we try to provide automated tools to design safe open systems. Open systems, which can be viewed as multi-agent systems, may be specified in ATL. The logic ATL has been especially introduced for that purpose. There exist two relevant extensions of ATL, namely ATL+ and ATL* (ATL+ being a restriction of ATL*).ATL+ allows Boolean combination of temporal operators, and ATL* also allows nesting of temporal operators. The tableau-based decision procedure for ATL is a constructive method to test the satisfiability of a given specification. It is constructive in the sense that it is possible to extract a model from the obtained tableau, whenever the root formula is indeed satisfiable. In this thesis, we propose two tableau-based decision procedures for ATL+ and ATL*, as well as an implementation of these procedures. Our procedures are sound, complete and optimal. Indeed, our two procedures run in 2Exptime. Up to our knowledge our implementation is the first running tool to decide satisfiability of both ATL and ATL* formulae. In the perspectives of this thesis, we discuss how it is possible to improve the extraction of models from tableaux for ATL, ATL+ and ATL*. We would like to obtain relatively small models at the end

Engineering Information and Communication Technology (ICT) for robust information sharing is the fundamental area of investigation in thesis. Robust workflow based information sharing systems have the potential to be part of robust information infrastructures providing positive effects for the individuals and teams as well as opportunities for societal and economical gains. Challenges in design and implementation of open socio-technical systems include identifying engineering principles empowering individual and team using the systems as well as supporting flexibility in design and maintenance. Of specific importance are principles supporting semantically correct information sharing. Information sharing in open socio-technical systems is given affordances due to coordination and exchange of services. Approaches ensuring robust semantically correct information sharing and user empowerments are key requirements especially since changes in context, roles and intentions are the rule and not the exception in socio-technical systems. Empirical observations of behaviours have been important for identifying critical patterns in workflow. A configuration of models and methods under the umbrella Participatory Design has been used including Ethnography and approaches based on Situation Theory, Knowledge Engineering, Interaction Design and Computer Supported Cooperative Work. The results of the configurations of methodologies are context sensitive since the methodologies are domain dependant. Three cases illustrating engineering support for empowerment of individuals and teams in open sociotechnical systems are presented. Two cases are based on studies performed in Sölvesborg and concerns engineering principles towards empowering individuals with cognitive impairments via ambient assistance. In the third case the focus is on hand-over situations and ontologies/abbreviations assuring semantically correct information sharing in distributed handling of critical emergency calls in Swedish Emergency Service Centres (SOS centres). The main contributions in this thesis, methodological contributions included, are engineering principles for open socio-technical systems from an empowerments perspective. The principles support understanding of workflows, information flows, interaction models, data models, semantics of information, trust, resilience, validation and training as well as assurance mechanisms in hand-over of critical operations. Identification and validation of key service qualities including mechanisms for improving performance critical tasks of semantics in information sharing are contributions. Service, Agent based and sensor approaches presented are final contributions.

Aunque los Sistemas Multiagente se suponen abiertos, la mayor parte de la investigación realizada se ha centrado en sistemas cerrados, diseñados por un sólo equipo de desarrollo, sobre un entorno homogéneo, y un único dominio.<br/>Esta tesis pretende avanzar hacia la consecución de Sistemas Multiagente abiertos. Nuestros esfuerzos se han centrado en desarrollar un marco de trabajo para Sistemas Multiagente que permita maximizar la reutilización de agentes en diferentes dominios, y soporte la formación de equipos bajo demanda, satisfaciendo los requerimientos de cada problema particular.<br/>Por un lado, este trabajo investiga el uso de Métodos de Solución de Problemas para describir las capacidades de los agentes con el objetivo de mejorar su reutilización. Hemos tenido que adaptar el modelo para trabajar con aspectos específicos de los agentes, como el lenguaje de comunicación y los protocolos de interacción.<br/>Por otro lado, esta tesis propone un nuevo modelo para el Proceso de Solución de Problemas Cooperativo, el cual introduce una fase de configuración previa a la formación de un equipo. El proceso de configuración se encarga de obtener un diseño de equipo, expresado en términos de las tareas a resolver, las capacidades a utilizar, y el conocimiento del dominio disponible.<br/>El marco de trabajo desarrollado ha sido puesto a prueba mediante la implementación de una infraestructura para agentes. Esta infraestructura proporciona un nivel de mediación social para los proveedores y clientes del sistema de resolución de problemas, sin imponer una arquitectura particular para los agentes participantes, ni un modelo mental o lógico para explicar la cooperación.<br/>Las contribuciones de este trabajo adoptan la forma de un marco de trabajo multi-capa, desde los conceptos más abstractos a los más concretos, para terminar con la implementación de una aplicación particular basada en agentes de información cooperativos.<br>Although Multi Agent Systems are supposed to be open systems, most of the initial research has focused on closed systems, which are designed by one developer team for one homogeneous environment, and one single domain.<br/>This thesis aims to advance some steps towards the realization of the open Multi Agent Systems vision. Our work has been materialized into a framework for developing Multi Agent Systems that maximize the reuse of agent capabilities across multiple application domains, and support the automatic, on-demand configuration of agent teams according to stated problem requirements.<br/>On the one hand, this work explores the feasibility of the Problem Solving Methods approach to describe agent capabilities in a way that maximizes their reuse. However, since Problem Solving Methods are not designed for agents, we have had to adapt them to deal with agent specific concepts concerning the agent communication languages and interaction protocols.<br/>One the other hand, this thesis proposes a new model of the Cooperative Problem Solving process that introduces a Knowledge Configuration stage previous to the Team Formation stage. The Knowledge Configuration process performs a bottom-up design of a team in term of the tasks to be solved, the capabilities required, and the domain knowledge available.<br/>The statements made herein are endorsed by the implementation of an agent infrastructure that has been tested in practice. This infrastructure has been developed according to the electronic institutions formalism to specifying open agent societies. This infrastructure provides a social mediation layer for both requesters and providers of capabilities, without imposing neither an agent architecture, nor an attitudinal theory of cooperation.<br/>The contributions of our work are presented as a multilayered framework, going from the more abstract aspects, to the more concrete, implementation dependent aspects, concluding with the implementation of the agent infrastructure and a particular application example for cooperative information agents.

COORDENAÇÃO DE APERFEIÇOAMENTO DO PESSOAL DE ENSINO SUPERIOR<br>A complexidade e a distribuição de sistemas multiagentes tornam seu desenvolvimento muito sujeito a falhas. Ao mesmo tempo, poucas iniciativas procuram resolver o problema através de teste automatizado da mesma maneira que no desenvolvimento orientado a objetos. Este trabalho apresenta um framework para o desenvolvimento de aplicações de teste automatizado através de agentes Stubs para o middleware M-Law. O middleware implementa um interpretador para a linguagem XMLaw, atuando como mediador da comunicação dos agentes segundo a definição das leis de interação.<br>Complexity and distribution of multi-agent systems make the development of such systems failure prone. In this context, few solutions try to solve the problem through automated tests as we see in object oriented development. This work presents a framework for test automated applications development through the use of Stub agents for the middleware M-Law. This middleware implements the law-enforcement mechanism for regulating OMAS specified by the description language XMLaw which implements a law definition language, called XMLaw, for agent interaction in open multi-agent systems.

PONTIFÍCIA UNIVERSIDADE CATÓLICA DO RIO DE JANEIRO<br>Governança trata heterogeneidade, autonomia e diversidade de interesses entre diferentes agentes em um sistema multi-agente (SMA), estabelecendo um conjunto de normas. A maioria dos mecanismos de aplicação de leis verifica a violação de normas apenas sob o aspecto da interação entre os agentes. Entretanto, em SMAs com a presença de agentes heterogêneos e independentemente projetados, existirão mensagens confidenciais, que só serão percebidas por seus remetentes e destinatários, e execução de ações que só serão observáveis pelos agentes que as executam ou por agentes que sofrem as conseqüências da execução dessas ações. Este trabalho apresenta uma abordagem que implementa um mecanismo de governança em SMA baseado em testemunhos. Agentes podem testemunhar fatos que estão relacionados à violação de normas as quais eles têm conhecimento. Este mecanismo é composto por três sub-sistemas: Reputação, julgamento e sanção. O foco deste trabalho está apenas no sub-sistema de julgamento, responsável por receber os testemunhos e prover decisões apontando se o agente realmente violou uma norma. Será mostrada a arquitetura deste sub-sistema e um processo de julgamento genérico. Finalmente, será ilustrada a utilização deste mecanismo através de um estudo de caso.<br>Governance copes with the heterogeneity, autonomy and diversity of interests among different agents in a multi-agent system (MAS) by establishing a set of norms. Most of the governance enforcement mechanisms usually check norm violations from the point of view of interaction protocols. However, in MAS, with the presence of heterogeneous and independently designed agents, there will be private messages, that will only be perceived by their senders and receivers, and execution of actions that will only be observed by the agents that execute them or by the agents that suffers from their consequences. This work presents a governance mechanism for MAS based on testimonies. Agents can witness facts that are related to norm violations witch they are aware of. The proposed mechanism is composed by three sub-systems: Reputation, judgment and sanction. This work focuses only in the judgment sub- system, witch is responsible for receiving testimonies and providing a decision, pointing out if an agent has really violated a norm. The judgment sub-system architecture and a generic judgment process will be presented. Finally, the use of this mechanism will be exemplified by a case study.

CONSELHO NACIONAL DE DESENVOLVIMENTO CIENTÍFICO E TECNOLÓGICO<br>Com o aumento da complexidade dos SMA´s governados por leis, tornouse mais difícil projetar e implementar requisitos referentes à tolerância a falhas, principalmente por não existir nenhum ponto de controle sobre tais sistemas. E, apesar disto, não são conhecidas soluções que abordem tais questões na literatura de governança de SMA´s baseados em leis. Contudo, existe uma técnica de tolerância a falhas, chamada Replicação de Agentes, que propõe reutilizar os conceitos de recuperação em sistemas distribuídos através da tolerância a falhas para a recuperação de agentes baseando- se em informações que identifiquem suas criticalidades. Entretanto, apesar de existirem algumas estratégias de estimativa de criticalidade de agentes, nenhuma apresenta o grau de granularidade suficiente para o monitoramento de SMA´s abertos governados por leis. Desta forma, o trabalho aqui proposto tem como objetivo apresentar um mecanismo de configuração de variação de criticalidade de agentes que interagem em SMA´s abertos. Para isto é proposta uma arquitetura para desenvolvimento de SMA´s abertos governados por leis tolerantes a falhas e uma abordagem para projetar tais mecanismos com simplicidade, modularidade e reuso. Este trabalho apresenta três contribuições principais: a extensão no modelo conceitual da linguagem declarativa de especificação de lei XMLaw com elementos para monitoramento da criticalidade dos agentes; a integração dos frameworks de governança de SMA´s abertos (M-Law) e de adaptação dinâmica de replicação de agentes (DIMAX); e a proposição de uma forma de documentação e derivação das leis. Por fim são apresentadas planilhas de resultados da aplicação do mecanismo proposto em dois estudos de caso.<br>With law-governed multi-agent systems complexity growth, it is harder to design and implement requirements related to dependability mainly because there is no control point over such systems. Despite the great importance of those problems, there is no solution that fully addresses those issues in the context of law-governed multi-agent systems. However, there is a fault tolerant technique called Agent Replication that proposes the reuse of some recovery concepts of distributed systems through a fault tolerance based approach on some data that helps the identification of critical agent. And, besides the existents strategies of agents´ criticality estimation, none of them presents the granularity degree necessary for the monitoring of law- governed open MAS´s. Due to it, the goal of this work is to present a mechanism of the configuration variation of the agents´ criticality that interacts in open MAS´s. To achieve this goal, an architecture to support the development of fault tolerant law-governed open MAS and an approach to design such mechanisms with simplicity, modularity and reuse were proposed. This work has three main contributions: an extension on the conceptual model of the law- specification declarative language XMLaw within new elements to monitor the agents´ criticality; the integration of two frameworks: one is for law-enforcement (M-Law), and the other one is for a dynamic and adaptative agents´ replication (DimaX); and a proposal of a means to document and derive the laws. Finally, we present some sheets of the results of the proposed mechanism application in two case studies.

COORDENAÇÃO DE APERFEIÇOAMENTO DO PESSOAL DE ENSINO SUPERIOR<br>O paradigma de agentes surgiu visando atender à necessidade de novas abstrações para o desenvolvimento de sistemas complexos e distribuídos. Para lidar com a mprevisibilidade do comportamento dos sistemas multi-agentes abertos, que são sistemas concorrentes e assíncronos formados por diversos agentes que agem com certo grau de autonomia e que podem interagir entre si para alcançar objetivos individuais, são usados mecanismos de governança na regulação das interações. Na maioria das abordagens existentes, a especificação das regras de governança é feita com o uso de linguagens declarativas ou de novas representações gráficas, o que pode tornar custosa essa tarefa e dificultar o uso desses mecanismos de governança. Esta dissertação apresenta a LawML, uma linguagem de modelagem baseada em UML para a especificação das regras de interação entre os agentes, com o objetivo de facilitar a tarefa de modelagem e, portanto, facilitar o uso de um mecanismo específico de governança baseado em leis de interação. Um conjunto de regras de transformação é apresentado junto com a linguagem, para permitir que os modelos gráficos de lei de interação sejam transformados em código no formato XMLaw - a linguagem declarativa do mecanismo de governança. Baseada nessas regras de transformação, é apresentada a ferramenta LawGenerator de transformação automática dos modelos de lei, para permitir o desenvolvimento das leis de interação com o foco nos modelos. E, por fim, esta abordagem é aplicada em um estudo baseado em um caso real de sistema distribuído com as características de um sistema multi-agente aberto - o SELIC, do Banco Central do Brasil.<br>The paradigm of agents appeared while aiming to satisfy the need for new abstractions for the development of complex and distributed systems. To manage with the unpredictable behavior of open multi-agent systems, governance mechanisms are used in the regulation of interactions between agents. This is due to the concurrent and asynchronous characteristics of these systems, which are formed by several agents who can act autonomically and can interact with each other to reach individual goals. In the majority of approaches, the governance rules are specified with declarative languages or new graphical representations, which can make this task costly and can make the use of these governance mechanisms difficult. This essay presents the LawML, a modeling language based on UML for the specification of rules for interactions between agents, which is aimed to facilitate the modeling task and, therefore, to facilitate the use of a specific governance mechanism based on interaction laws. A set of transformation rules is presented in addition to the language to allow the graphical interaction law models to be transformed into the declarative language of the governance mechanism, the XMLaw format code. To allow the model-driven development of interaction laws, it is presented the LawGenerator, a tool for the automatic transformation of the law model, based on these transformation rules. Finally, this approach is applied to a case study based on a real distributed system, the Brazilian Central Bank SELIC system, with the characteristics of an open multi-agent system.

Open distributed computing applications are becoming increasingly commonplace nowadays. In many cases, these applications are composed of multiple autonomous agents, each with its own aims and objectives. In such complex systems, communication between these agents is usually essential for them to perform their task, to coordinate their actions and share their knowledge. However, successful and meaningful communication can only be achieved by a shared understanding of each other's messages. Therefore efficient mechanisms are needed to reach a mutual understanding when exchanging expressions from each other's world model and background knowledge. We believe the de facto mechanisms for achieving this are ontologies, and this is the area explored in this thesis [88]. However, supporting shared understanding mechanisms for open distributed applications is a major research challenge. Specifically, one consequence of a system being open is the heterogeneity of the agents. Agents may have conflicting goals, or may be heterogeneous with respect to their beliefs or their knowledge. Forcing all agents to use a common vocabulary defined in one or more shared ontologies is, thus, an oversimplified solution, particularly when these agents are designed and deployed independently of each other. This thesis proposes a novel approach to overcome vocabulary heterogeneity, where the agents dynamically negotiate the meaning of the terms they use to communicate. While many proposals for aligning two agent ontologies have been presented in the literature as the current standard approaches to resolve heterogeneity, they are lacking when dealing with important features of agents and their environment. Motivated by the hypothesis that ontology alignment approaches should reflect the characteristics of autonomy and rationality that are typical of agents, and should also be tailored to the requirements of an open environment, such as dynamism, we propose a way for agents to define and agree upon the semantics of the terms used at run-time, according to their interests and preferences. Since agents are autonomous and represent different stakeholders, the process by which they come to an agreement will necessarily only come through negotiation. By using argumentation theory, agents generate and exchange different arguments, that support or reject possible mappings between vocabularies, according to their own preferences. Thus, this work provides a concrete instantiation of the meaning negotiation process that we would like agents to achieve, and that may lead to shared understanding. Moreover, in contrast to current ontology alignment approaches, the choice of a mapping is based on two clearly identified elements: (i) the argumentation framework, which is common to all agents, and (ii) the preference relations, which are private to each agent. Despite the large body of work in the area of semantic interoperabiJity, we are not aware of any research in this area that has directly addressed these important requirements for open Multi-Agent Systems as we have done in this thesis. Supplied by The British Library - 'The world's knowledge'

Agent-oriented programming paradigm and mobile software agents are getting the attention of IT community because they look a promising solution for the development of systems that fit the computational requirements of emerging mobile computing, ubiquitous computing and pervasive computing. Software agents consist of autonomous processes that run on a hosting platform and that are aimed at achieving specific owners' goals they were programmed for. Agents can migrate from host to host in order to exploit resources and services that are available in the distributed system. In the 90's, at the beginning of the agent model definition and development, agent systems addressed applications with strictly local perspectives: the whole system was contained within a defined boundary and was composed of a defined set of known agents and hosts. Agent systems are now moving towards open perspective applications: agent platforms are involving in wide size and higly dynamic distributed systems without well-defined boundaries. Security issues that arise in open agent systems require new protection mechanisms that combine the protection from malicious agents with local boundaries that are open to the community of users. Soft security mechanisms are based on the employment of trust and reputation concepts that are derived from the same definitions used in social relationships. Soft security is employed in trust-based decision making: it support the evaluation of risks that are involved in interactions with unknown agents; reputation information informs agents about the behavior and honesty of agents in the community according to the opinions coming from experienced interactions. We propose a model for reputation management that is compliant with the characteristics of the agent systems (e.g. dynamicity of agents), and that meets also requirements that arise in these systems in critical situations, e.g. low computational power in small portable devices that host agents and low and unstable communication bandwidth in mobile devices. The model is based on the evaluation of opinions that are collected within context groups. A context group is a coalition of agents that provide opinions related to the behaviors of agents. In a context group opinions are related to a specific context. Hence the proposed model introduces a context feature to reputation information. The reputation model provides agents with information that regards both the general fairness of an agent and its behavior related to the context where it is rated. The model is also composed of a network of distributed informative points that are aimed at supporting trust information sharing among context groups. The model is consistent for what concerns both the agent model and the agent social perspective because its design adopted an agent-centric approach but at the same time it adopted solutions in order to employ the model in large society of agents.<br>Il paradigma di programmazione orientato agli agenti e gli agenti software stanno ottenendo l'attenzione della comunità dell'Information Technology in quanto sembrano essere una soluzione promettente per lo sviluppo di sistemi che si possano applicare alle discipline emergenti del mobile computing, ubiquitous computing e pervasive computing. Gli agenti software consistono in processi autonomi che vengono eseguiti in piattaforme host, ed hanno l'obiettivo di raggiungere gli scopi per i quali i proprietari li hanno programmati. Gli agenti hanno la capacità di migrare da un host ad un altro sfruttando le risorse e i servizi che sono disponibili nel sistema distribuito. Negli anni novanta, agli inizi della definizione del modello ad agenti, i sistemi ad agenti erano utilizzati per applicazioni che avevano prospettive strettamente locali: tutto il sistema era contenuto all'interno di frontiere ben definite ed era composto di un definito insieme di agenti e risorse di calcolo tra di loro reciprocamente noti. I sistemi ad agenti stanno ora ampliando il loro utilizzo in applicazioni con prospettive più ampie: le piattaforme ad agenti vengono utilizzate in sistemi distribuiti caratterizzati da ampie dimensioni e da un'alta dinamicità senza confini ben definiti. I problemi di sicurezza che sorgono nei sistemi aperti ad agenti richiedono nuovi meccanismi di protezione che combinino la protezione da agenti maliziosi con le frontiere dei sistemi locali aperte alla vasta comunità di potenziali utilizzatori. I sistemi di sicurezza definiti soft sono basati sull'impiego dei concetti di fiducia e reputazione, i quali sono derivati dalle stesse definizioni usate comunemente nelle relazioni sociali. La sicurezza soft è utilizzata nelle decisioni basate sul grado di fiducia: essa è impiegata nella valutazione dei rischi che possono derivare da interazioni con agenti non conosciuti. Le reputazione rendono danno informazione agli agenti sul tipo di comportamento e il grado di onestà di altri agenti nella comunità, in base alle opinioni che derivano da interazioni avvenute precedentemente. Viene proposto un modello per la gestione della reputazione che sia coerente con le caratteristiche di un sitema ad agenti (per esempio la dinamicità degli agenti) e che rispetti anche i requisiti che sorgono in questi sistemi in particolari situazioni critiche, quali per esempio capacita' computazionali ridotte dei dispositivi portatili sui quali vengono eseguiti gli agenti e comunicazione instabile in dispositivi mobili. Il modello si basa sulla valutazione di opinioni che vengono raccolte all'interno di gruppi di contesto. Un gruppo di contesto è una coalizione di agenti i quali forniscono opinioni riguardanti il comportamento degli agenti. In un gruppo di contesto le opinioni sono legate allo specifico contesto. Il modello proposto quindi introduce la caratteristica di contesto nelle informazioni di reputazione. Il modello di reputazione fornisce agli agenti informazioni che riguardano sia l'onestà generale di un agente, sia il suo comportamento rispetto il contesto dove esso viene valutato. Il modello si compone anche di una rete di punti di informazione distribuiti che hanno lo scopo di supportare la distribuzione delle informazioni fra gruppi di contesto. Il modello è consistente sia dal punto di vista del modello agente sia da quello della sua dimensione sociale, in quanto è stato disegnato attorno al concetto di agente ma allo stesso tempo sono state adottate soluzioni che permettano il suo utilizzo in ampie società di agenti.

In this study, the researcher created a model and software application for identifying the factors that are relevant in the decision making process to select and implement Open-source applications in higher education. Open-source applications provide the programming syntax to the user for customization. Unlike proprietary software, where the source code is unavailable and illegal to alter, an Open-Source Software (OSS) application authorizes the user to edit and recompile the application to meet the specific needs of the institution or organization. OSS applications are either free or purchasable for a one-time fee. The rising cost of proprietary software has motivated many academic institutions to consider implementing OSS. Many IT professionals are investigating the advantages and disadvantages of open-source applications in an attempt to mitigate expensive yearly fees, licensures and maintenance costs required by proprietary software vendors. The data collected in the study represented OSS and non-OSS enabled institutions that were members of the Council of Higher Education and Accreditation (CHEA) Organization. Of the data collected from the institutions, a portion of the information was set aside for validation purposes. The model created in this research addressed the OSS concerns in higher education by identifying the experiences, institutional characteristics, and technical systems relevant to the selection and implementation of OSS applications. The researcher used the Visual Basic .NET programming language to create the model and software application that provided academic institutions with technical OSS information and support. The Intelligent Software Agent (ISA) simplified the data analysis process by providing a Graphical User Interface (GUI) for the user to enter and receive data. The results of this research allowed institutions to specify certain criteria such as highest degree awarded, relevant characteristics, and technological factors and then receive implementation suggestions for adopting OSS applications. The validation process indicated that the tabled data in the model and generated suggestions of the ISA were statistically comparable with the data that was set-aside for validation purposes.

PONTIFÍCIA UNIVERSIDADE CATÓLICA DO RIO DE JANEIRO<br>CONSELHO NACIONAL DE DESENVOLVIMENTO CIENTÍFICO E TECNOLÓGICO<br>Um dos desafios de desenvolvimento de software é produzir aplicativos que são projetados para evoluir reduzindo esforços de manutenção. Diversas técnicas desenvolvidas para a governança de leis de interação em sistemas multiagentes abertos foram propostas, no entanto a flexibilidade e a reutilização de leis não ocorrem de forma sistemática com estas técnicas. A tecnologia de gframeworks visa orientar o projeto e a implementação de leis de interação em sistemas multiagentes abertos, com o objetivo de produzir mecanismos de governança de leis de interação. A flexibilidade em g- frameworks é obtida através da introdução de incrementos específicos que as instâncias em desenvolvimento requerem, de modo a completar e adaptar as funcionalidades originais do g-framework. A reutilização em g-frameworks vem justamente do re-aproveitamento de um mesmo projeto e código de lei de interação em instâncias geradas a partir do g-framework. Os benefícios obtidos por tal abordagem podem impactar positivamente o desenvolvimento de software em termos do custo e tempo total de construção de uma família de mecanismos de governança de sistemas multiagentes. Para isto, são apresentadas técnicas de governança de sistemas multiagentes abertos e técnicas de reutilização de leis de interação. Um método de orientação é proposto para guiar o desenvolvimento de g-frameworks. Experimentos foram desenvolvidos e são descritos neste documento.<br>One of the challenges of software development is to produce applications that are designed to evolve, reducing maintenance efforts. Many techniques developed to govern the interaction laws in open multi- agent systems were proposed, but the flexibility and reuse concerns of interaction laws were not systemically fulfilled by them. The technology of g- frameworks intends to guide the design and the implementation of interaction laws in open multi-agent systems, aiming to facilitate the production of interaction law governance mechanisms. The flexibility in g-frameworks is achieved by specific increments that the instances under development require, to complete and adapt the original functionalities of the g-framework. The reuse in g- frameworks is related to a common design and codification of that interaction laws that are shared by instances developed with the g-framework. The benefits of this approach might positively impact the development of software considering the costs and the necessary time to construct the family of governance mechanisms of multiagent systems. In this thesis, some techniques to promote reuse of interaction laws were propose to fulfill this goal. One method to orient the development of g-frameworks is proposed. Experiments were developed and they are described in this thesis.

Cette thèse traite du problème où une équipe d'agents coopératifs et autonomes, évoluant dans un environnement stochastique partiellement observable, et œuvrant à la résolution d'une tâche complexe, doit modifier dynamiquement sa composition durant l'exécution de la tâche afin de s'adapter à l'évolution de celle-ci. Il s'agit d'un problème qui n'a été que peu étudié dans le domaine de la planification multi-agents. Pourtant, il existe de nombreuses situations où l'équipe d'agent mobilisée est amenée à changer au fil de l'exécution de la tâche.Nous nous intéressons plus particulièrement au cas où les agents peuvent décider d'eux-même de quitter ou de rejoindre l'équipe opérationnelle. Certaines fois, utiliser peu d'agents peut être bénéfique si les coûts induits par l'utilisation des agents sont trop prohibitifs. Inversement, il peut parfois être utile de faire appel à plus d'agents si la situation empire et que les compétences de certains agents se révèlent être de précieux atouts.Afin de proposer un modèle de décision qui permette de représenter ces situations, nous nous basons sur les processus décisionnels de Markov décentralisés et partiellement observables, un modèle standard utilisé dans le cadre de la planification multi-agents sous incertitude. Nous étendons ce modèle afin de permettre aux agents d'entrer et sortir du système. On parle alors de système ouvert. Nous présentons également deux algorithmes de résolution basés sur les populaires méthodes de recherche arborescente Monte-Carlo. Le premier de ces algorithmes nous permet de construire des politiques jointes séparables via des calculs de meilleures réponses successives, tandis que le second construit des politiques jointes non séparables en évaluant les équipes dans chaque situation via un système de classement Elo. Nous évaluons nos méthodes sur de nouveaux jeux de tests qui permettent de mettre en lumière les caractéristiques des systèmes ouverts<br>This thesis addresses the problem where a team of cooperative and autonomous agents, working in a stochastic and partially observable environment towards solving a complex task, needs toe dynamically modify its structure during the process execution, so as to adapt to the evolution of the task. It is a problem that has been seldom studied in the field of multi-agent planning. However, there are many situations where the team of agents is likely to evolve over time.We are particularly interested in the case where the agents can decide for themselves to leave or join the operational team. Sometimes, using few agents can be for the greater good. Conversely, it can sometimes be useful to call on more agents if the situation gets worse and the skills of some agents turn out to be valuable assets.In order to propose a decision model that can represent those situations, we base upon the decentralized and partially observable Markov decision processes, the standard model for planning under uncertainty in decentralized multi-agent settings. We extend this model to allow agents to enter and exit the system. This is what is called agent openness. We then present two planning algorithms based on the popular Monte-Carlo Tree Search methods. The first algorithm builds separable joint policies by computing series of best responses individual policies, while the second algorithm builds non-separable joint policies by ranking the teams in each situation via an Elo rating system. We evaluate our methods on new benchmarks that allow to highlight some interesting features of open systems

The relaxation of international boundaries has enabled the globalisation of markets making available an ever increasing number of specialised suppliers and markets. Inevitably this results in supply chains sharing suppliers and customers reflected in a network of relationships. Within this context firms buyers configure their supply relationships based on their perception of supply risk. Risk is managed by either increasing trust or commitment or by increasing the number of suppliers. Increasing trust and commitment facilitates collaboration and reduces the propensity for a supplier to exit the relationship. Conversely, increasing the number of suppliers reduces dependency and increases the ease of making alternative supply arrangements. The emergent network of relationships is dynamic and complex, and due in no small part to the influence of inventory management practices, tightly coupled. This critical organization of the network describes a system that contrary to existing supply chain conceptualisation exists far from equilibrium, requiring a different more appropriate theoretical lens through which to view them. This thesis adopts a Complex Adaptive Systems (CAS) perspective to position supply networks as tightly coupled complex systems which according to Normal Accident Theory (NAT) are vulnerable to disruptions as a consequence of normal operations. The consequential boundless and emergent nature of supply networks makes them difficult to research using traditional empirical methods, instead this research builds a generalised supply network agent based computer model, allowing network constituents (agents) to take autonomous parallel action reflecting the true emergent nature of supply networks. This thesis uses the results from a series of carefully designed computer experiments to elucidate how supply networks respond to a variety of market structures and permitted agent behaviours. Market structures define the vertical (between tier) and horizontal (within tier) levels of price differentiation. Within each structure agents are permitted to autonomously modify their prices (constrained by market structure) and collaborate by sharing demand information. By examining how supply networks respond to different permitted agent behaviours in a range of market structures this thesis makes 4 contributions. Firstly, it extends NAT by incorporating the adaptive nature of supply network constituents. Secondly it extends supply chain management by specifying supply networks as dynamic not static phenomena. Thirdly it extends supply chain risk management through developing an understanding of the impact different permitted behaviour combinations on the networks vulnerability to disruptions in the context of normal operations. Finally by developing the understanding how normal operations impact a supply networks vulnerability to disruptions it informs the practice of supply chain risk management.

In a complex adaptive system, diverse agents perform various actions without adherence to a predefined structure. The achievement of collaborative actions will be the result of continual interactions among them that shape a dynamic network. Agents may form an ad hoc organization based on the dynamic network of interactions for the purpose of achieving a long-term objective, which we termed a Network Organization (NO). Fervent and agile communication on social networking sites provides opportunities for potential issues to trigger individuals into individual actions as well as the attraction and mobilization of like-minded individuals into an NO that is both physically and virtually emergent. Examples are the rapid pace of Arab Spring proliferation and the diffusion rate of the Occupy Movement. We are motivated by a spontaneously formed NO as well as the quality of plasticity that enables the organization to change rapidly to describe an NO. Thus, we present a paradigm that serves as a reference model for organizations of socially networked individuals. This paradigm suggests modular components that can be combined to form an ad hoc network organization of agents. We touch on how this model accounts for external change in an environment through internal adjustment. For the predominant influences of the network substrate in an NO, multiple effects of it have an impact on the NO behaviors and directions. We envisioned several dimensions of such effects to include synergy, social capital, externality, influence, etc. A special focus in this work is measuring synergy and social capital as two predominant network effects. Synergy is perceived as different modalities of compatibility among agents when performing a set of coherent and correspondingly different actions. When agents are under no structural obligation to contribute, synergy is quantified through multiple forms of serendipitous agent chosen benevolence among them. The approach is to measure four types of benevolence and the pursuant synergies stemming from agent interactions. Social capital is another effect of networking that describes the accumulation of positive values of social flow and perceived trust plus abundance of communication over the common topic of NO. We provide measurement of social capital based on an agents’ expected benevolence. We examine those two effects in two different case studies — one case of a virtual organization and another of a real world terrorist organization — that best illustrate the main tenets of our conceptualization.

Peu importe l’échelle d’observation, les systèmes biologiques allant des molécules aux cellules et les organismes multicellulaires aux communautés, manifeste des comportements collectifs. De nombreuses idées explicatives ont été avancées basé sur la perception locale. Dans chacun de ces cas, des progrès ont été possible parce que le phénotype au niveau collectif est évident et observable à l’oeil nu, mais aussi en raison de la fonction d’objectif visée dans la compréhension de ces comportements. Ici, j’ai construit un hybride machine-écosystème qui implique un dispositif d’observation, couplé à un générateur de paysage lumineux dans une boucle qui contient un processus d’apprentissage à chaque étape en fonction des données capturées de l’écosystème observé. Afin de combler la difficulté expérimentale entre modélisation et réalité, compte tenu de la complexité profonde du système étudié car la modélisation peut alors ´évoluer avec le la matière observée, qui est réactive envers le paysage adaptatif qui va interagir avec lui, cela permettant d’atteindre des horizons inexplorés et non expérimentalement implémentables pour le moment. Mon projet comporte plusieurs étapes. Construction de l’hybride. Développement et mise en pratique d’algorithmes d’interaction. Mise en place de populations microbiennes simples qui peuvent être maintenu et dont le comportement peut être manipulé via la machine. Expériences pour démontrer le principe<br>No matter the scale of observation, biological systems ranging from molecules to cells, and multicellular organisms to communities, manifest collective behaviours. Many explanatory ideas have been put forward based on local perception. In each of these cases progress has been possible because the collective-level phenotype is obvious and observable to the naked eye, but also because of the objective function targeted in the understanding of these behaviours. Here I constructed a machine-ecosystem hybrid that involve an observation device, coupled to a light landscape generator in an automated loop that contain a learning process at every step based on data captured from the ecosystem being observed. In order to break the reality gap between modelisation and reality, taking into account the deep complexity of the studied system as the modelisation can then co evolve with the observed matter, that is reactive toward the adaptive landscape that interact with it, and then reach unexplored and not experimentally implementable fields of heuristic. My project involves several stages. Construction of the hybrid. Application and development interaction algorithms. Establishment of simple microbial populations that can be sustained and whose behaviour can be manipulated via the machine. Experiments to demonstrate proof of principle

Sensor networks are being used in several emerging applications not even imagined some years ago due to advances in sensing, computing, and communication techniques. However, these advances also pose various challenges that must be faced. One important challenge is related to the autonomous capability needed to setup and adapt the networks, which decentralizes the control of the network, saving communication and energy resources. Middleware technology helps in addressing this kind of problem, but there is still a need for additional solutions, particularly considering dynamic changes in users' requirements and operation conditions. This paper presents an agent-based framework acting as an integral part of a middleware to support autonomous setup and adaptation of sensor networks. It adds interoperability among heterogeneous nodes in the network, by means of autonomous behavior and reasoning. These features also address the needs for system setup and adaptations in the network, reducing the communication overhead and decentralizing the decision making mechanism. Additionally, preliminary results are also presented.

Lo scopo principale di questa ricerca è quello di inquadrare le possibili relazioni causali fra corruzione negli approvvigionamenti pubblici e performance delle amministrazioni locali. A tal fine, viene adottato un disegno di ricerca che integra dinamicamente metodi quantitativi e qualitativi in ogni fase del processo di studio. Gli approcci Dynamic Performance Management (DPM) e Governance (DPG), supportati da interviste approfondite non-strutturate, modellazioni formali e simulazioni quantitative, sono qui adottati per analizzare un caso di studio rappresentativo di un piccolo comune italiano in cui si sono verificati alcuni episodi di corruzione nelle attività di approvvigionamento nei primi anni 2000. In particolare, l'Ente locale in questione è stato sciolto due volte per infiltrazioni di stampo mafioso e, attualmente, versa in una situazione di dissesto finanziario. Il lavoro analizza i possibili esiti di tali fatti sulla performance organizzativa dell'Ente nel suo complesso sulla base di tre fonti: dati primari qualitativi generati da interviste frontali convergenti non-strutturate; dati secondari estrapolati da sentenze passate in giudicato e da archivi ad accesso aperto; ed un'ampia rassegna della letteratura. All'inizio, una panoramica della letteratura aiuta il lettore a comprendere i contenuti, le teorie ed i confini della corruzione. Successivamente, viene proposto un esame delle strategie di misurazione e delle misure più diffuse per prevenirla e contrastarla. Nel complesso, una particolare attenzione è riservata agli approvvigionamenti nei contesti pubblici locali. A seguito di una discussione sui possibili vantaggi e svantaggi, in termini di opportunità e deterrenze corruttive, derivanti dall'adozione dei paradigmi di governance nel settore pubblico più diffusi, gli approcci DPM e DPG vengono analizzati per comprendere il loro contributo teorico nel supportare i responsabili politici ed i manager ad arginare i fenomeni di corruzione. Successivamente, sulla base di alcune tecniche di codifica applicate alle interviste frontali non-strutturate svolte con alcuni Pubblici Ufficiali, un approccio esplorativo-descrittivo del caso-studio selezionato consente di comprendere la misura in cui gli eventi corruttivi investigati in questa sede abbiano inciso nel tempo sulla performance complessiva del Comune preso in esame. In seguito, viene adottata una prospettiva sistemica e dinamica di performance management per inquadrare le relazioni di causa-effetto emergenti dal caso di studio. L'assunzione di un approccio DPM consente ai politici ed ai dirigenti pubblici di progettare, implementare e valutare strategie anticorruzione fattibili, efficaci ed efficienti a livello di governo locale. Più precisamente, l'utilizzo, in un grafico DPM, di driver di performance adeguati al rischio di corruzione legato alle sue cause strutturali ed individualistiche può porre rimedio non soltanto alle riconosciute ambiguità e carenze derivanti dall'adozione di “bandiere rosse” nei processi di approvvigionamento pubblico, ma anche ai fallimenti dei controlli direzionali meccanicistici nel rilevare l'effettiva presenza di corruzione, fornendo ai decisori tempestivi segnali derivanti dagli effetti deleteri prodotti da siffatte pratiche clandestine. Inoltre, enfatizzare il ruolo della moralità civica comunitaria a livello di sistema può supportare la comprensione di alcuni risultati controintuitivi della passata ricerca sulla corruzione negli approvvigionamenti pubblici e dedurre in che modo gli investimenti nelle tecnologie dell'informazione e della comunicazione (TIC) e la formazione del personale possano migliorare responsabilità e competenza dei governi locali. Per quanto riguarda la professionalità dei dirigenti e dello staff quale causa individualistica di corruzione, il patronage politico derivante da opportunità legali risulta, per il caso di studio, significativo nello spiegare il cattivo andamento degli approvvigionamenti pubblici nel tempo. All'interno di questo quadro, la visualizzazione DPM consente altresì di distinguere più dettagliatamente la corruzione dallo spreco di risorse dovuto ad azioni non corruttive. In sintesi, un approccio DPM può consentire ai manager pubblici di mantenere i loro “radar cognitivi” costantemente (re)attivi, in modo da: identificare e sopprimere pratiche illecite negli approvvigionamenti pubblici; rilevare illeciti emergenti che potrebbero essere trascurati dai tradizionali approcci diagnostici e interattivi del controllo direzionale; favorire l'apprendimento etico e migliorare il valore pubblico generato. Infine, la prospettiva DPG può essere efficace nel supportare la formulazione e la messa a punto di strategie anticorruzione basate sulla collaborazione multistakeholder, nonché nel sondarne la fattibilità e gli impatti nel tempo all'interno di aree locali caratterizzate da strutture di governance carenti. Pertanto, sulla base di precedenti progetti di successo realizzati da governi locali di diverse parti del mondo, nell'ultima parte di questa tesi di ricerca viene utilizzato un approccio DPG per delineare e valutare una strategia anticorruzione per il caso in esame secondo un'impostazione di collaborative governance, così da inquadrare possibili sinergie ed interdipendenze tra i soggetti rilevanti, quali leve critiche per contrastare il rischio sistemico di corruzione a livello locale.<br>The main purpose of this research is to frame the possible causal relationships between corruption in public procurement and performance of local governments. With this aim, a fully-integrated research design is adopted to dynamically mix quantitative and qualitative methods at every phase of the research process. The Dynamic Performance Management (DPM) and Governance (DPG) approaches, supported by in-depth interviews, formal modelling and simulations, are here adopted to analyse a representative case-study of an Italian small Municipality, where some corruption episodes in procurement activities occurred in the early 2000s. In particular, the local Authority in question was disbanded twice for mafia-like infiltrations and is currently facing a financial instability. In the light of that, the overall work studies the possible outcomes of those facts on the organisational performance as a whole, based on three sources: qualitative primary data generated by face-to-face convergent interviews; secondary data retrieved from both documents describing legal cases and open-access repositories; an extensive literature review. At the beginning, a broad and composite literature overview helps the reader become aware about the contents, theories and boundaries of corruption. Thereafter, an examination of the most widespread measurement strategies and measures to either prevent or repress it is proposed. Overall, a special focus is set on procurement in local public contexts. Following a discussion on the possible advantages and disadvantages of the most common public sector paradigms in terms of opportunities and constraints for corruption, the DPM and DPG views are explored to understand their theoretical contribution in supporting policy- and decision-makers to curb corruption phenomena in heterogeneous governance contexts. Afterwards, based on coding techniques, an exploratory-descriptive approach of the selected case-study allows for a better understanding of the extent to which the investigated corruption events impacted on the overall performance of the Municipality under scrutiny over time, by means of non-structured face-to-face interviews held with some Public Officials in 2019. Therefore, a system perspective in performance management is adopted to frame the emerging cause-and-effect relationships of the case-study. Assuming a DPM approach allows politicians and public managers to design, implement and assess feasible, effective and efficient anti-corruption strategies at local government level. More precisely, the use in a DPM chart of performance drivers adjusted for the risk of corruption linked to its structural and individualistic causes may not only put right to ambiguities and flaws deriving from the adoption of ‘red-flags’ in public procurement, but also to failures of mechanistical controls in detecting the actual presence of corruption, thus providing decision-makers with prompt signals arising from the emergent effects of clandestine practices. In addition, emphasising the role of community civic morality at system level may back up the understanding of some counterintuitive results in the past research on corruption in public procurement and deduce to what extent investments in Information and Communication Technologies (ICTs) and personnel training may enhance local government accountability and expertise. With regard to managers’ and staff’s professionalism, as individualistic cause of corruption, political patronage stemming from legal opportunities results significant in explaining poor performance in public procurement over time. Within this framework, the DPM view also allows for better singling corruption out from resource waste due to non-corrupt actions. In summary, a DPM approach may lead public managers to constantly maintain their ‘cognitive radar’ reactive, so as to identify and suppress unlawful practices in procurement, detect emerging malfeasances that could be otherwise overlooked by traditional static diagnostical and interactive approaches, foster ethical learning and enhance community outcomes. Finally, the DPG perspective may be effective in supporting formulation and fine-tuning of collaboration-based anti-corruption strategies and probing their feasibility and impacts over time within local areas characterised by poor governance structures. Hence, on the basis of previous successful projects throughout the world, in the last part of this research thesis a DPG approach is used to outline and evaluate an anti-corruption strategy for the case under scrutiny according to collaboration settings, in a way to frame possible synergies and interdependencies among relevant participants as critical levers to hinder systemic risk of corruption at local level.

Les communautés virtuelles sont des systèmes sociotechniques dans lesquels des entités (humaines et/ou artificielles) répartis à travers le monde se réunissent autour d’intérêts et/ou d’objectifs communs. Afin de réaliser ces objectifs, les membres de la communauté doivent collaborer en partageant leurs ressources et/ou connaissances. Or, toute collaboration comporte une part de risque dans la mesure où les membres peuvent se comporter de manière non coopérative ou malveillante. Dans de tels contextes, où les mécanismes de sécurité standard ne suffissent plus, la confiance est rapidement devenue un facteur déterminant lors de la prise de décision. Le travail présenté dans cette thèse s’attaque à la problématique de la gestion de la confiance dans les communautés virtuelles ouvertes et décentralisées. Pour cela, nous avons proposé une infrastructure de gestion de la confiance adaptative et conforme socialement (ASC-TMS). L’aspect novateur de ce système réside dans sa faculté à exhiber des propriétés sociales et adaptatives. L’aspect social du ASC-TMS fait référence à la capacité de notre système à prendre des décisions qui soient sûres non seulement pour l’individu mais également et surtout pour les autres membres de la communauté. Par ailleurs, l’aspect adaptatif du système fait référence à la capacité du système à prendre des décisions qui soient en parfaite adéquation avec l’environnement dans lequel ces décisions sont prises. Ainsi, cette thèse constitue une nouvelle étape vers l’automatisation de l’évaluation de la confiance en assistant les membres des communautés virtuelles ouvertes et décentralisées dans leur prise de décision. Le système a été implémenté et déployé en utilisant la plateforme de développement multi-agent JaCaMo. Nous avons également illustré l’applicabilité de notre approche sur un scénario réel de communauté virtuelle d’innovation ouverte. Enfin, nous avons évalué notre système expérimentalement en utilisant la plateforme de simulation multi-agent Repast. Les résultats obtenus montrent que l’utilisation de notre système avait un impact positif sur la dynamique des communautés dans lesquels il est a été utilisé<br>Virtual communities (VCs) are socio-technical systems wherein distributed individuals (human and/or artificial) are grouped together around common objectives and goals. In such systems, participants are massively collaborating with each other’s by sharing their private resources and knowledge. A collaboration always bears the risk that one partner exhibits uncooperative or malicious behaviour. Thus, trust is a critical issue for the success of such systems. The work presented in this dissertation addresses the problem of trust management in open and decentralised virtual communities (VCs). To address this problem, we proposed an Adaptive and Socially-Compliant Trust Management System (ASC-TMS). The novelty of ASC-TMS lies in its ability to exhibit social-awareness and context-awareness features. Social-awareness refers to the ability of the trust management system (TMS) to handle the social nature of VCs by making trust evaluations that are collectively harmful, while context-awareness refers to the ability of the system to handle the dynamic nature of VCs by making trust evaluations that are always in adequacy with the context in which these evaluations are undertaken. Thus, the contributions made in this thesis constitute an additional step towards the automation of trust assessment. We provided accordingly a novel trust management system that assists members of open and decentralised virtual communities in their trust decisions. The system has been implemented and deployed using the JaCaMo multi-agent platform. We illustrated also the applicability of on a real life open innovation virtual community scenario. Finally, the ASC-TMS has been experimentally evaluated using the multi-agent based Repast simulation platform. The preliminary results show that the use of our system significantly improves the stability of the virtual communities in which it has been deployed

Les approches habituelles pour la protection de la privacité s'attachent à définir un niveau de sensibilité pour chaque information. Cette information est soit publique, soit privée et sa circulation est restreinte à un groupe d'agents prédéfini. Dans cette thèse, nous nous appuyons sur la théorie de l'intégrité contextuelle, qui propose de redéfinir la notion de violation de privacité. Selon cette théorie, toute transmission peut déclencher une violation de privacité suivant le contexte dans lequel elle a lieu. Cette thèse utilise la théorie de l'intégrité contextuelle afin de proposer un modèle de protection de la privacité pour les systèmes multi-agents décentralisés: le modèle PrivaCIAS. Afin de contrôler les agents dans le système, le modèle PrivaCIAS fournit un ensemble de normes qui permet la mise en place d'un contrôle social basé sur la confiance. Le modèle donne le contrôle aux agents pour constater les violations (selon l'intégrité contextuelle), puis punir les contrevenants en les excluant du système sans avoir besoin de recourir à une autorité centrale. Ce modèle vise les réseaux sociaux décentralisés comme champ d'application.

Cette thèse présente le modèle d'un mécanisme de raisonnement social fondé sur la théorie de la dépendance. Ce modèle permet à un agent de raisonner sur autrui et plus particulièrement de calculer ses relations et situations de dépendance. Un agent est dépendant d'un autre si celui-ci peut l'aider/l'empêcher d'atteindre un de ses buts. Nous considérons notre mécanisme de raisonnement social comme un composant essentiel pour la conception d'agents artificiels réellement autonomes, évoluant dans un univers multi-agents ouvert. La notion d'ouverture désigne la capacité d'ajouter ou de retirer dynamiquement dans le système des agents. Comme dans ces systèmes l'organisation des agents ne peut pas être spécifiée pendant la phase de conception, la résolution coopérative de problèmes est fondée sur la formation dynamique de coalitions. Dans ce contexte, des agents doivent être capables de s'adapter aux changements dynamiques du système, en particulier en évaluant pendant la phase de résolution si leurs buts sont réalisables et si leurs plans sont exécutables. Comme nous ne supposons pas que les agents soient bienveillants, notre modèle fournit un critère pour évaluer les partenaires le plus susceptibles d'accepter une proposition de coalition. Enfin, comme dans ces systèmes des agents n'ont pas généralement une représentation complète et correcte les uns des autres, notre modèle leur permet de détecter une inconsistance au niveau de la société et de choisir un contexte à être maintenue. Nous avons implémenté ce mécanisme de raisonnement social en utilisant une programmation orientée objet. Nous l'avons utilisé pour développer deux applications, le simulateur DEPNET et le système DEPINT, qui illustrent son utilisation selon deux perspectives scientifiques différentes. D'une part, selon une perspective de simulation sociale, notre modèle fournit un outil informatique permettant l'analyse et la prédiction des divers schémas intéressants d'interaction sociale, et l'évaluation du pouvoir social des agents. D'autre part, selon une perspective de résolution de problèmes, notre modèle peut être utilisé pour concevoir dynamiquement l'organisation des agents dans un contexte de systèmes multi-agents ouverts<br>This thesis presents the model of a social reasoning mechanism based on dependence theory. This model enables an agent to reason about the others, in particular to calculate his dependence relations and dependence situations. An agent is said to be dependent on another if the latter can help/prevent him to achieve one of his goals. We consider our social reasoning mechanism as an essential building block for the design of really autonomous artificial agents, which are immersed in an open multi-agent world. By open, we mean that agents may enter or leave the agency at any moment. In such systems, as the organisation of the agents can not be conceived at design time, the cooperative problem solving paradigm is based on dynamic coalition formation. In this context, agents must be able to adapt themselves to dynamically changing conditions, by evaluating at execution time if their goals are achievable and if their plans are feasible. As we do not suppose that agents are benevolent, our model proposes a criterion to evaluate which partners are more susceptible to accept a proposition of coalition. Finally, as in these kind of systems agents usually do not have a complete and correct representation of each other, our model helps them to detect an agency level inconsistency and to choose a context to be maintained. We have implemented our social reasoning mechanism using an object-oriented approach, and we have used it to develop two applications, the DEPNET simulator and the DEPINT system, which illustrate respectively its usage in two different scientific perspectives. On one hand, concerning social simulation, our model provides a computational tool for the analysis and prediction of the occurrence of several interesting patterns of social interactions, and for the evaluation of the agents' social power. On the other hand, with respect to problem solving, our model can be used to design dynamic agents' organizations in a context of open multi-agent systems

碩士<br>逢甲大學<br>資訊電機工程碩士在職專班<br>93<br>A mobile agent system can be used to distribute computation throughout a network environment. Most of the mobile agents rely on a location service system to communicate with each other in the network environment. The location service system must provide reliable location information and guarantee delivery messages. Aglets is a popular Java-based mobile agent system and was originally developed at the IBM Tokyo Research Laboratory. The Aglets system is distributed under the IBM Public License, then release to sourceforge.net as an Open Source project in 2000. However, the complete location service solutions are not currently supported by the Aglets. Our research is based on the IBM Public License and location service technology to design and implement an open sourced location service system for the Aglets mobile agent system. Our location service system has four major advantages: easy to integrate for legacy applications, easy to join servers in the region, easy to link among multi-regions for different laboratories, and easy to implement in your code that make its range of applications more extensive. Finally, the experimental results demonstrate that our system is efficient in various network environments.

abstract: For this master's thesis, an open learner model is integrated with Quinn, a teachable robotic agent developed at Arizona State University. This system is represented as a feedback system, which aims to improve a student’s understanding of a subject. It also helps to understand the effect of the learner model when it is represented by performance of the teachable agent. The feedback system represents performance of the teachable agent, and not of a student. Data in the feedback system is thus updated according to a student's understanding of the subject. This provides students an opportunity to enhance their understanding of a subject by analyzing their performance. To test the effectiveness of the feedback system, student understanding in two different conditions is analyzed. In the first condition a feedback report is not provided to the students, while in the second condition the feedback report is provided in the form of the agent’s performance.<br>Dissertation/Thesis<br>Masters Thesis Computer Science 2016

碩士<br>聖約翰科技大學<br>資訊與通訊系碩士班<br>107<br>The thesis focuses on integrating the open data of the New Taipei City Government and global positioning system technology to build an intelligent traffic assistant agent system with location-based service (LBS) and linked open data (LOD) technology. The proposed system mainly stores data into the database one by one and analyzes those data through the government's open materials for providing more convenient instant traffic information to users and government related units. Taking the traffic information of the New Taipei City Government as an example, the system can not only display the optimal size through smart handheld devices (including IOS and Android systems) and computer devices, but also integrate database analysis to instantly view traffic information. Users don’t need to download and install the app and just open the browser and directly keyin the URL to enter the system to implement the system cross-platform services, and save a lot of system development costs. The system also provides a user return function. When the vehicle sensor is not installed in some road sections, but the vehicle is blocked in the road section, the user can report the latest road condition information through the user's timely return. After comparing and analyzing the system with some related applications, although the functions are similar, the proposed system has more practicality, information accessibility and excellent user experience.

碩士<br>聖約翰科技大學<br>資訊與通訊系碩士班<br>104<br>This thesis aims to integrate the linked open library (Linked Open Data, LOD) technology and Global Positioning System (GPS) technology to build and explore the related technologies of agent systems for restaurant information services. Given an example on the open data platform of the New Taipei City Government, the proposed system can integrate Google Maps to provide and check the timely, complete, correct and full-service restaurant information with the user's query by the Government approved, and then select the corresponding position services. Research and development of the proposed system wish that the user can not only choose the safty stores recomended and certificated by the open data platform, but also go through the way of location-based services (LBS) to provide users with related services around the target restaurant. The proposed method in information itself not only can allow users to feel at ease, the proposed system but also can provide users with the most useful information block to further reach the differentiation between the presented system and relevantly existent information systems. The preliminary system presentation and comparison show the feasibility of the proposed system.

One of the major problems that discourages the development of Decision Support Systems (DSSs) is the un-standardised DSS environment. Computers that support modern business processes are no longer stand-alone systems, but have become tightly connected both with each other and their users. Therefore, having a standardised environment that allows different DSS applications to communicate and cooperate is crucial. The integration difficulty is the most crucial problem that affects the development of DSSs. Therefore, an open and standardised environment for integrating various DSSs is required. Despite the critical need for an open architecture in the DSS designs, the present DSS architectural designs are unable to provide a fundamental solution to enhance the flexibility, connectivity, compatibility, and intelligence of a DSS. The emergence of intelligent agent technology fulfils the requirements of developing innovative and efficient DSS applications as intelligent agents offer various advantages, such as mobility, flexibility, intelligence, etc., to tackle the major problems in existing DSSs. Although various agent-based DSS applications have been suggested, most of these applications are unable to balance manageability with flexibility. Moreover, most existing agent-based DSSs are based on agent-coordinated design mechanisms, and often overlook the living environment for agents. This could cause the difficulties in cooperating and upgrading agents because the agent-based coordination mechanisms have limited capabilities to provide agents with relatively comprehensive information about global system objectives. This thesis proposes a novel multi-agent-based architecture for DSS, called Agentbased Open Connectivity for Decision support systems (AOCD). The AOCD architecture adopts a hybrid agent network topology that makes use of a unique feature called the Matrix-agent connection. The novel component, i.e. Matrix, provides a living environment for agents; it allows agents to upgrade themselves through interacting with the Matrix. This architecture is able to overcome the difficulties in concurrency control and synchronous communication that plague many decentralised systems. Performance analysis has been carried out on this framework and we find that it is able to provide a high degree of flexibility and efficiency compared with other frameworks. The thesis explores the detailed design of the AOCD framework and the major components employed in this framework including the Matrix, agents, and the unified Matrices structure. The proposed framework is able to enhance the system reusability and maximize the system performance. By using a set of interoperable autonomous agents, more creative decision-making can be accomplished in comparison with a hard-coded programmed approach. In this research, we systematically classified the agent network topologies, and developed an experimental program to evaluate the system performance based on three different agent network topologies. The experimental results present the evidence that the hybrid topology is efficient in the AOCD framework design. Furthermore, a novel topological description language for agent networks (TDLA) has been introduced in this research work, which provides an efficient mechanism for agents to perceive the information about their interconnected network. A new Agent-Rank algorithm is introduced in the thesis in order to provide an efficient matching mechanism for agent cooperation. The computational results based on our recently developed program for agent matchmaking demonstrate the efficiency and effectiveness of the Agent-Rank algorithm in the agent-matching and re-matching processes

One of the major problems that discourages the development of Decision Support Systems (DSSs) is the un-standardised DSS environment. Computers that support modern business processes are no longer stand-alone systems, but have become tightly connected both with each other and their users. Therefore, having a standardised environment that allows different DSS applications to communicate and cooperate is crucial. The integration difficulty is the most crucial problem that affects the development of DSSs. Therefore, an open and standardised environment for integrating various DSSs is required. Despite the critical need for an open architecture in the DSS designs, the present DSS architectural designs are unable to provide a fundamental solution to enhance the flexibility, connectivity, compatibility, and intelligence of a DSS. The emergence of intelligent agent technology fulfils the requirements of developing innovative and efficient DSS applications as intelligent agents offer various advantages, such as mobility, flexibility, intelligence, etc., to tackle the major problems in existing DSSs. Although various agent-based DSS applications have been suggested, most of these applications are unable to balance manageability with flexibility. Moreover, most existing agent-based DSSs are based on agent-coordinated design mechanisms, and often overlook the living environment for agents. This could cause the difficulties in cooperating and upgrading agents because the agent-based coordination mechanisms have limited capabilities to provide agents with relatively comprehensive information about global system objectives. This thesis proposes a novel multi-agent-based architecture for DSS, called Agentbased Open Connectivity for Decision support systems (AOCD). The AOCD architecture adopts a hybrid agent network topology that makes use of a unique feature called the Matrix-agent connection. The novel component, i.e. Matrix, provides a living environment for agents; it allows agents to upgrade themselves through interacting with the Matrix. This architecture is able to overcome the difficulties in concurrency control and synchronous communication that plague many decentralised systems. Performance analysis has been carried out on this framework and we find that it is able to provide a high degree of flexibility and efficiency compared with other frameworks. The thesis explores the detailed design of the AOCD framework and the major components employed in this framework including the Matrix, agents, and the unified Matrices structure. The proposed framework is able to enhance the system reusability and maximize the system performance. By using a set of interoperable autonomous agents, more creative decision-making can be accomplished in comparison with a hard-coded programmed approach. In this research, we systematically classified the agent network topologies, and developed an experimental program to evaluate the system performance based on three different agent network topologies. The experimental results present the evidence that the hybrid topology is efficient in the AOCD framework design. Furthermore, a novel topological description language for agent networks (TDLA) has been introduced in this research work, which provides an efficient mechanism for agents to perceive the information about their interconnected network. A new Agent-Rank algorithm is introduced in the thesis in order to provide an efficient matching mechanism for agent cooperation. The computational results based on our recently developed program for agent matchmaking demonstrate the efficiency and effectiveness of the Agent-Rank algorithm in the agent-matching and re-matching processes

Chrysanthos Dellarocas and Mark Klein.<br>Title from cover. "May 2000."<br>Includes bibliographical references (p. 13-16).<br>Supported in part by NSF. IIS-9803251 Supported in part by DARPA. F30602-98-2-0099

Multi-agent systems are applied to distributed problem-solving applications because of their ability to overcome the limitations that individual agents face when solving complex problems. Large numbers of agents acting as problem-solvers on networks suggest a virtual marketplace. In this marketplace, groups of self-interested agents can interact to solve highly constrained and distributed problems by assuming varying roles and forming “temporary teams”. This dissertation presents a decision making mechanism for multi-agent team formation between self-interested agents in a competitive, open and dynamic environment. An agent perceives environmental uncertainties, and models those uncertainties into simplified categories such as risks and benefits. The dissertation further demonstrates how an agent’s attitudes shape how risk and rewards are weighted when making decisions among multiple alternatives. Accordingly, agent-borne attitudes toward proactive behavior, risk, reward, and urgency are proposed as the basis of the proposed team formation mechanism. Finally, a learning technique assists an agent in continuously learning what attitudes it needs in order to adapt to dynamic environments and increase its resulting rewards.<br>text

Various computer-based dispatching systems have been developed for managing truck and shovel pairing in surface mining operations and are being used to varying degrees of success. Systems have shown the ability to increase production and maintain ore quality within prescribed upper and lower limits provided there is a stable operational environment. However, operational environments in mining are uncertain and highly variable. Upsets such as equipment breakdowns, or changing weather conditions often occur and no claims have been made about the success of these systems to react to these upsets and successfully adapt to new operational conditions generated by upsets. In an ant colony different activities are performed simultaneously by specialized individuals. However when the environment with an ant colony changes or experiences a major upset, the configuration of task allocations within the colony change to adapt to the new conditions. In this thesis, the task allocation model developed for ant colonies was modified and used to develop a dispatch algorithm, the Agent Based Model, which reacts reliably to changes and upsets in surface mining operations. The algorithm was simulated over a twelve-hour shift using AUTOMOD®, a discrete event simulation program. The simulation results of the Agent Based Model are compared to that of the Fixed Assignment Method used by current dispatch systems in which each truck is permanently assigned to a particular shovel. The total production of ore and waste from the Agent Based Model is consistently greater than that from the Fixed Assignment Method because an Ant Based system allows re-assignment (task reallocation). The simulations also show that the Agent Based system reliably adapts and limits the impact of upsets on a mining operation.

<p>This dissertation examines the impact of active shooters upon patrons attending large outdoor events. There has been a spike in shooters targeting densely populated spaces in recent years, to include open-air venues. The 2019 Gilroy Garlic Festival was selected for modeling replication using AnyLogic software to test various experiments designed to reduce casualties in the event of an active shooter situation. Through achievement of validation to produce identical outcomes of the real-world Gilroy Garlic Festival shooting, the researcher established a reliable foundational model for experimental purposes. This active shooter research project identifies the need for rapid response efforts to neutralize the shooter(s) as quickly as possible to minimize casualties. Key findings include the importance of armed officers patrolling event grounds to reduce response time, the need for adequate exits during emergency evacuations, incorporation of modern technology to identify the shooter’s location, and applicability of a 1:548 police to patron ratio.</p>