Showing papers by "Munindar P. Singh published in 2009"

Operators for propagating trust and their evaluation in social networks

Abstract: Trust is a crucial basis for interactions among parties in large, open systems. Yet, the scale and dynamism of such systems make it infeasible for each party to have a direct basis for trusting another party. For this reason, the participants in an open system must share information about trust. However, they should not automatically trust such shared information. This paper studies the problem of propagating trust in multiagent systems. It describes a new algebraic approach, shows some theoretical properties of it, and empirically evaluates it on two social network datasets. This evaluation incorporates a new methodology that involves dealing with opinions in an evidential setting.

Amoeba: A methodology for modeling and evolving cross-organizational business processes

Abstract: Business service engagements involve processes that extend across two or more autonomous organizations. Because of regulatory and competitive reasons, requirements for cross-organizational business processes often evolve in subtle ways. The changes may concern the business transactions supported by a process, the organizational structure of the parties participating in the process, or the contextual policies that apply to the process. Current business process modeling approaches handle such changes in an ad hoc manner, and lack a principled means for determining what needs to be changed and where. Cross-organizational settings exacerbate the shortcomings of traditional approaches because changes in one organization can potentially affect the workings of another.This article describes Amoeba, a methodology for business processes that is based on business protocols. Protocols capture the business meaning of interactions among autonomous parties via commitments. Amoeba includes guidelines for (1) specifying cross-organizational processes using business protocols, and (2) handling the evolution of requirements via a novel application of protocol composition. This article evaluates Amoeba using enhancements of a real-life business scenario of auto-insurance claim processing, and an aerospace case study.

Commitment-Based Service-Oriented Architecture

Abstract: Existing service-oriented architectures are formulated in terms of low-level abstractions far removed from business services. In a new SOA, the components are business services and the connectors are patterns, modeled as commitments, that support key elements of service engagements.

Multiagent commitment alignment

Abstract: Commitments provide a basis for understanding interactions in multiagent systems. Successful interoperation relies upon the interacting parties being aligned with respect to their commitments. However, alignment is nontrivial in a distributed system where agents communicate asynchronously and make different observations. We propose a formalization for commitments that ensures alignment despite asynchrony. This formalization consists of three elements: (1) a semantics of commitment operations; (2) messaging patterns that implement the commitment operations; and (3) weak constraints on agents' behaviors to ensure the propagation of vital information. We prove that our formalization ensures alignment. We illustrate the generality of our formalization with several real-life scenarios.

Choice, interoperability, and conformance in interaction protocols and service choreographies

Abstract: Many real-world applications of multiagent systems require independently designed (heterogeneous) and operated (autonomous) agents to interoperate. We consider agents who offer business services and collaborate in interesting business service engagements. We formalize notions of interoperability and conformance, which appropriately support agent heterogeneity and autonomy. With respect to autonomy, our approach considers the choices that each agent has, and how their choices are coordinated so that at any time one agent leads and its counterpart follows, but with initiative fluidly shifting among the participants. With respect to heterogeneity, we characterize the variations in the agents' designs, and show how an agent may conform to a specification or substitute for another agent. Our approach addresses a challenging problem with multi-party interactions that existing approaches cannot solve. Further, we introduce a set of edit operations by which to modify an agent design so as to ensure its conformance with others.

Modelling Interactions via Commitments and Expectations

Abstract: Organizational models often rely on two assumptions: openness and heterogeneity. This is, for instance, the case with organizations consisting of individuals whose behaviour is unpredictable, whose internal structure is unknown, and who do not necessarily share common goals, desires, or intentions. This fact has motivated the adoption of social-based approaches to modelling interaction in organizational models. The idea of social semantics is to abstract away from the agent internals and provide a social meaning to agent message exchanges. In this chapter, we present and discuss two declarative, social semantic approaches for modelling interaction. The first one takes a state-oriented perspective, and models interaction in terms of commitments. The second one adopts a rule-oriented perspective, and models interaction in terms of logical formulae expressing expectations about agent interaction. We use a simple interaction protocol taken from the e-commerce domain to present the functioning and features of the commitment- and expectation-based approaches, and to discuss various forms of reasoning and verification that they accommodate, and how organizational modelling can benefit from them.

Business Modeling via Commitments

Abstract: Existing computer science approaches to business modeling offer low-level abstractions such as data and control flows, which fail to capture the business intent underlying the interactions that are central to real-life business models. In contrast, existing management science approaches are high-level but not only are these semiformal, they are also focused exclusively on managerial concerns such as valuations and profitability. This paper proposes a novel business metamodel based on commitments that considers additional agent-oriented concepts, specifically, goals and tasks. It proposes a set of business patterns and algorithms for checking model completeness and verification of agent interactions. Unlike traditional models, our approach marries rigor and flexibility, providing a crisp notion of correctness and compliance independent of specific executions.

Correctness properties for multiagent systems

Abstract: What distinguishes multiagent systems from other software systems is their emphasis on the interactions among autonomous, heterogeneous agents. This paper motivates and characterizes correctness properties for multiagent systems. These properties are centered on commitments, and capture correctness at a high level. In contrast to existing approaches, commitments underlie key correctness primitives understood in terms of meaning; for example, commitment alignment maps to interoperability; commitment discharge maps to compliance. This paper gives illustrative examples and characterizations of these and other properties. The properties cover the specification of the principal artifacts—protocols, roles, and agents—of an interaction-based approach to designing multiagent systems, and thus provide the formal underpinnings of the approach.

Enhancing Tropos with Commitments

Abstract: This paper motivates a novel metamodel and methodology for specifying cross-organizational business interactions that is based on Tropos. Current approaches for business modeling are either high-level and semiformal or formal but low-level. Thus they fail to support flexible but rigorous modeling and enactment of business processes. This paper begins from the well-known Tropos approach and enhances it with commitments. It proposes a natural metamodel based on commitments and a methodology for specifying a business model. This paper includes an insurance industry case study that several researchers have previously used.

Programming multiagent systems without programming agents

Abstract: We consider the programming of multiagent systems from an architectural perspective. Our perspective emphasizes the autonomy and heterogeneity of agents, the components of multiagent systems, and focuses on how to specify their interconnections in terms of high-level protocols. In this manner, we show how to treat the programming of a multiagent system as an architectural endeavor, leaving aside the programming of individual agents who might feature in a multiagent system as a secondary concern.

Information Sharing among Autonomous Agents in Referral Networks

Abstract: Referral networks are a kind of P2P system consisting of autonomous agents who seek and provide services, or refer other service providers. Key applications include service discovery and selection, and knowledge sharing. An agent seeking a service contacts other agents to discover suitable service providers. An agent who is contacted may autonomously ignore the request or respond by providing the desired service or giving a referral. This use of referrals is inspired by human interactions, where referrals are a key basis for judging the trustworthiness of a given service. The use of referrals differentiates such networks from traditional P2P information sharing systems, which are based on request flooding. Not only does the use of referrals enable an agent to control how its request is processed, it also provides an architectural basis for four kinds of interaction policies. InterPol is a language and framework supporting such policies. InterPol provides an ability to specify requests with hard and soft constraints as well as a vocabulary of application-independent terms based on interaction concepts. Using these, InterPol enables agents to reveal private information and accept others' information based on subtle relationships. In this manner, InterPol goes beyond traditional referral and other P2P systems in supporting practical applications. InterPol has been implemented using a Datalog-based policy engine for each agent. It has been applied on scenarios from a (multinational) health care project. The contribution of this paper is in a general referrals-based architecture for information sharing among autonomous agents, which is shown to effectively capture a variety of privacy and trust requirements of autonomous users.

Architecture for Affective Social Games

Abstract: The importance of affect in delivering engaging experiences in entertainment and education is well recognized. We introduce the Koko architecture, which describes a service-oriented middleware that reduces the burden of incorporating affect into games and other entertainment applications. Koko provides a representation for affect, thereby enabling developers to concentrate on the functional and creative aspects of their applications. The Koko architecture makes three key contributions: (1) improving developer productivity by creating a reusable and extensible environment; (2) yielding an enhanced user experience by enabling independently developed applications to collaborate and provide a more coherent user experience than currently possible; (3) enabling affective communication in multiplayer and social games.

Elements of a business-level architecture for multiagent systems

Abstract: Existing architectures for multiagent systems emphasize low-level messaging-related considerations. As a result, the programming abstractions they provide are also low level. In recent years, commitments have been applied to support flexible interactions among autonomous agents. We present a layered multiagent system architecture based on commitments. In this architecture, agents are the components, and the interconnections between the agents are specified in terms of commitments, thus abstracting away from low level details. A crucial layer in this architecture is a commitment-based middleware that plays a vital role in ensuring interoperation and provides commitment-related abstractions to the application programmer. Interoperation itself is defined in terms of commitment alignment. This paper details various aspects of this architecture, and shows how a programmer would write applications to such an architecture.

Enhancing Tropos with Commitments A Business Metamodel and Methodology

Abstract: This paper motivates a novel metamodel and methodology for specifying cross-organizational business interactions that is based on Tropos. Current approaches for business modeling are either high-level and semiformal or formal but low-level. Thus they fail to support flexible but rigorous modeling and enactment of business processes. This paper begins from the well-known Tropos approach and enhances it with com- mitments. It proposes a natural metamodel based on commitments and a methodology for specifying a business model. This paper includes an insurance industry case study that several researchers have previously used.

Formal methods in agent-oriented software engineering

Abstract: There is a growing interest among agent and multiagent system developers for formal methods. Formal methods are means to define and realize correct specifications of multiagent system. The benefits of formal methods become clearer when we recognize the cost of developing a defective multiagent system. This paper seeks to introduce engineers to the possibilities of applying formal methods for multiagent systems. To this end, it discusses selected formal methods approaches for multiagent systems for which there is tool support. These works have been organized into two broad categories: those formal methods constituting a development method in themselves and those intended to complement an existing development method.

Integrating marine observatories into a system-of-systems: Messaging in the US Ocean Observatories Initiative

Abstract: The Ocean Observatories Initiative (OOI) will implement ocean sensor networks covering a diversity of oceanic environments, ranging from the coastal to the deep ocean. Construction will begin in Fall 2009, with deployment phased over five years. The integrating feature of the OOI is a comprehensive Cyberinfrastructure (CI), whose design is based on loosely-coupled distributed services, and whose elements are expected to reside throughout the physical components; from seafloor instruments to autonomous vehicles to deep sea moorings to shore facilities to computing and storage infrastructure. The OOI-CI provides novel capabilities for data acquisition, distribution, modeling, planning and interactive control of oceanographie experiments. The architecture comprises six subsystems: four elements address the oceanographie science- and education-driven operations of the OOI integrated observatory, and two elements provide core infrastructure services for the distributed, message-based, service-oriented integration and communication infrastructure, as well as the virtualization of computational and storage resources. All OOI functional capabilities and resources represent themselves as services to the observatory network, with precisely defined service access protocols based on message exchange. This paper presents an overview of the OOI services and focuses on the strategy for service-oriented integration and the publish-subscribe model for communication.

Methodology for engineering affective social applications

Abstract: Affective applications are becoming increasingly mainstream in entertainment and education. Yet, current techniques for building such applications are limited, and the maintenance and use of affect is in essence handcrafted in each application. The Koko architecture describes middleware that reduces the burden of incorporating affect into applications, thereby enabling developers to concentrate on the functional and creative aspects of their applications. Further, Koko includes a methodology for creating affective social applications, called Koko-ASM. Specifically, it incorporates expressive communicative acts, and uses them to guide the design of an affective social application. With respect to agentoriented software engineering, Koko contributes a methodology that incorporates expressives. The inclusion of expressives, which are largely ignored in conventional approaches, expands the scope of AOSE to affective applications.

Evidence-based trust in distributed agent systems

Abstract: Trust is a crucial basis for interactions among parties in large, open distributed systems. Yet, the scale and dynamism of such systems make it infeasible for each party to have a direct basis for trusting another party. For this reason, the participants in an open system must share information about trust. Traditional models of trust employ simple heuristics and ad hoc formulas, without adequate mathematical justification. These models fail to properly address the challenges of combining trust from conflicting sources, dealing with malicious agents, and updating trust. This dissertation understands an agent Alice's trust in an agent Bob in terms of Alice's certainty in her belief that Bob is trustworthy. Unlike previous approaches, this dissertation formulates certainty in terms of a statistical measure defined over a probability distribution of the probability of positive outcomes. Specifically this dissertation makes the following contributions. It (1) Develops a mathematically well-formulated approach for an evidence-based account of trust; proves desirable properties of certainty; and establishes a bijection between evidence and trust. (2) Defines a concatenation, an aggregation, and a selection operator to propagate trust, and proves desirable properties of these operators. (3) Develops trust update mechanisms and formally analyzes their properties. (4) Extends the definition of certainty from binary events to multivalued events. Establishes a bijection between Dempster-Shafer belief space and evidence space, and defines a novel combination operator, which is commutative and associative. In contrast with traditional combination operators, which ignore conflict and sometimes yield counterintuitive results, the proposed operator treats conflict naturally.

Koko: architecture and methodology for engineering social affective applications

Abstract: Representing and reasoning about affect (e.g. emotion and mood) is essential for producing believable characters and empathic interactions with users, both of which are necessary for effective entertainment and education. Leading applications of interest include pedagogical tools [1], military training simulations [2], and educational games [3].

Commitment alignment as a basis for interoperability in service engagements

Abstract: The expanding interest in the modeling and enactment of service engagements creates fresh challenges for software engineering. Existing work on “technical” services, such as on web and grid services, bears little relationship to real-life “business” services that we consider here. Conventional software engineering is ill-equipped to deal with distributed information systems comprised of autonomous entities representing the business partners involved in a typical service engagement. Although many have recognized the importance of communication in the modeling and enactment of distributed systems, existing approaches treat communication at a low level. Specifically, existing messagingbased approaches consider the flow, but not the meanings, of the messages. Thus they can facilitate interoperation only at a correspondingly low level. In contrast, we give primacy to the meanings of the messages exchanged, based on the notion of commitments. Doing so yields a new high-level notion of interoperability, which we term alignment. We propose a distributed (locally executable) method that guarantees the alignment of business partners despite asynchrony.

Formal aspects of classifying and selecting business services

Abstract: This paper uses the term "service" for a service instance, and the term "agent" for an agent who provides or consumes a service We consider real-life or business services (eg, business process out-sourcing, software development service involving human experts) We distinguish business from computational (eg, Web or grid) services based on the fact that business services lack the typical input-output structure of computational services For example, one can model a temperature service as one that takes a zipcode as input and produces the current temperature as output By contrast, it would not help to model a software development service as one that takes a "business problem" as input and produces a suite of "software modules" as output First, it is clearly beyond the scope of current practice to create formal classes or a type system of business problems and software modules Second, business services are not invoked but are engaged, and would rarely take single-shot inputs and produce single-shot outputs Third, business service providers would offer a continuum of expertise along which they can provide effective services For example, a provider who is good at payroll management may also be able to provide retirement plan management, in contrast with the temperature service example above, which has no other function Fourth, the selection of business services relies on the agents' evaluation of previous engagements

Formal Aspects of Classifying and Selecting Business Services (Extended Abstract)

Abstract: This paper uses the term “service” for a service instance, and the term “agent” for an agent who provides or consumes a service. We consider real-life or business services (e.g., business process outsourcing, software development service involving human experts). We distinguish business from computational (e.g., Web or grid) services based on the fact that business services lack the typical inputoutput structure of computational services. For example, one can model a temperature service as one that takes a zipcode as input and produces the current temperature as output. By contrast, it would not help to model a software development service as one that takes a “business problem” as input and produces a suite of “software modules” as output. First, it is clearly beyond the scope of current practice to create formal classes or a type system of business problems and software modules. Second, business services are not invoked but are engaged, and would rarely take single-shot inputs and produce single-shot outputs. Third, business service providers would offer a continuum of expertise along which they can provide effective services. For example, a provider who is good at payroll management may also be able to provide retirement plan management, in contrast with the temperature service example above, which has no other function. Fourth, the selection of business services relies on the agents’ evaluation of previous engagements. Service selection inherently considers instances (runtime), not types (design-time). A service can be judged along many different dimensions in an empirical basis. Since agents are autonomous,

Koko: engineering affective applications

Abstract: Representing and reasoning about affect (i.e. emotion) is essential for producing believable characters and empathic interactions with users, both of which are necessary for effective entertainment and education. Leading applications of interest include pedagogical tools [1], military training simulations [2], and educational games [3].