Showing papers presented at "European Workshop on Multi-Agent Systems in 2006"

Bidding Languages and Winner Determination for Mixed Multi-unit Combinatorial Auctions.

Abstract: We introduce a new type of combinatorial auction that allows agents to bid for goods to buy, for goods to sell, and for transformations of goods. One such transformation can be seen as a step in a production process, so solving the auction requires choosing the sequence in which the accepted bids should be implemented. We introduce a bidding language for this type of auction and analyse the corresponding winner determination problem.

Exploiting the Environment for Coordinating Agent Intentions.

Abstract: One large and quite interesting family of MAS applications is characterized (1) by their large scale in terms of number of agents and physical distribution, (2) by their very dynamic nature and (3) by their complex functional and non-functional requirements. This family includes a.o. manufacturing control, traffic control and web service coordination. BDI-based agent architectures have proven their usefulness in building MASs for complex systems - their explicit attention for coping with dynamic environments is one obvious explanation for this. For the family of applications mentioned above, the complexity of the software for the individual agents using traditional BDI-approaches, however, is overwhelming. In this paper, we present an innovative approach to BDI agents which alleviates agent complexity through so-called "delegate MASs", which use the environment and its resources to obtain BDI functionality. Delegate MASs consist of light-weight agents, which are issued either by resources for building and maintaining information on the environment, or by task agents in order to explore the options on behalf of the agents and to coordinate their intentions. We describe the approach, and validate it in a case study of manufacturing control. The evaluation in this case study shows the feasibility of the approach in coping with the large scale of the application and shows that the approach elegantly achieves flexibility in highly dynamic environments.

DimaX: a fault-tolerant multi-agent platform

Abstract: Fault tolerance is an important property of large-scale multi-agent systems as the failure rate grows with both the number of the hosts and deployed agents, and the duration of computation. Several approaches have been introduced to deal with some aspects of the fault-tolerance problem. However, most existing solutions are ad hoc. Thus, no existing multi-agent architecture or platform provides a fault-tolerance service that can be used to facilitate the design and implementation of reliable multi-agent systems. So, we have developed a fault-tolerant multi-agent platform (named DimaX). DimaX deals with fail-stop failures like bugs and/or breakdown machines. It brings fault-tolerance for multi-agent applications by using replication techniques. It is based on a replication framework (named DARX).

Developing and Evaluating a Generic Metamodel for MAS Work Products.

Abstract: MAS development requires an appropriate methodology. Rather than seek a single, ideal methodology, we investigate the applicability of method engineering, which focuses on project-specific methodology construction from existing method fragments and provides an appealing approach to organize, appropriately access and effectively harness the software engineering knowledge of MAS methodologies. In this context, we introduce a generic metamodel to serve as a representational infrastructure to unify the work product component of MAS methodologies. The resultant metamodel does not focus on any class of MAS, nor does it impose any restrictions on the format of the system requirements; rather, it is an abstraction of how the work product elements in any MAS are structured and behave both at design time and run-time. Furthermore, in this paper we validate this representational infrastructure by analysing two well-known existing MAS metamodels. We sketch how they can be seen as subtypes of our generic metamodel, providing early evidence to support the use of our metamodel towards the construction of situated MAS methodologies.

Learning Automata as a Basis for Multi Agent Reinforcement Learning.

Abstract: In this paper we summarize some important theoretical results from the domain of Learning Automata. We start with single stage, single agent learning schema's, and gradually extend the setting to multistage multi agent systems. We argue that the theory of Learning Automata is an ideal basis to build multi agent learning algorithms.

Model Checking Abilities under Incomplete Information Is Indeed Delta2-complete.

Abstract: We study the model checking complexity of Alternating-time temporal logic with imperfect information and imperfect recall (atlir). Contrary to what we have stated in [11], the problem turns out to be ∆2 -complete, thus con rming the initial intuition of Schobbens [18]. We prove ∆2 -hardness through a reduction of the SNSAT problem, while the membership in ∆ P 2 stems from the algorithm presented in [18].

Agent-based Architecture in An Ambient Intelligence Context.

Abstract: This paper provides an insight on the special requirements of accessibility content and services in an ambient intelligence context and proposes an agent-based approach into a more general service oriented architecture for addressing them. It is based on previous approaches for agent-based information systems regarding infomobility services adding the mobility impaired people special requirements. The result is an architecture with multiple agents on the user’s nomad device that address the ambient intelligence issue and a family of dedicated personal assistance agents to each type of mobility impairment whose deliberation provides the best solutions for people that have a combination of impairments. In this paper we pay a particular attention on the task of integrating the multi-agent system in the overall architecture.

Reinforcement Learning based Control of Traffic Lights in Non-stationary Environments: A Case Study in a Microscopic Simulator.

Abstract: Coping with dynamic changes in traffic volume has been the object of recent publications. Recently, a method was proposed, which is capable of learning in non-stationary scenarios via an approach to detect context changes. For particular scenarios such as the traffic control one, the performance of that method is better than a greedy strategy, as well as other reinforcement learning approaches, such as Q-learning and Prioritized Sweeping. The goal of the present paper is to assess the feasibility of applying the above mentioned approach in a more realistic scenario, implemented by means of a microscopic traffic simulator. We intend to show that to use of context detection is suitable to deal with noisy scenarios where non-stationarity occurs not only due to the changing volume of vehicles, but also because of the random behavior of drivers in what regards the operational task of driving (e.g. deceleration probability). The results confirm the tendencies already detected in the previous paper, although here the increase in noise makes the learning task much more difficult, and the correct separation of contexts harder.

Sellers Competing for Buyers in Online Markets: Reserve Prices, Shill Bids, and Auction Fees.

Abstract: We consider competition between sellers offering similar items in concurrent online auctions through a mediating auction institution, where each seller must set its individual auction parameters (such as the reserve price) in such a way as to attract buyers. We show that in the case of two sellers with asymmetric production costs, there exists a pure Nash equilibrium in which both sellers set reserve prices above their production costs. In addition, we show that, rather than setting a reserve price, a seller can further improve its utility by shill bidding (i.e., bidding as a buyer in its own auction). This shill bidding is undesirable as it introduces inefficiencies within the market. However, through the use of an evolutionary simulation, we extend the analytical results beyond the two-seller case, and we then show that these inefficiencies can be effectively reduced when the mediating auction institution uses auction fees based on the difference between the auction closing and reserve prices.

Coordinating Tasks in Agent Organizations.

Abstract: Support for new forms of organization and social interaction requires understanding the influence of structure on behavior. Goal dependencies indicate some relationship between roles, through which actions can be coordinated. Social relationships determine different types of power links between roles. Efficient coordination requires that goal dependency and power structure are well tuned to each other. In this paper, we will investigate what is the exact nature of this relationship between roles in an organization and what are the consequences of different structure forms. We will also see what is the difference if the relations are not hierarchical but organized through a market or network structure.

Multi-agent Coordination and Anticipation Model to Design a Road Traffic Simulation Tool.

Abstract: The use of a multi-agent approach in the design of a traffic simulation tool is innovative since most of actual tools are still based on mathematical approach. In this paper we present a multi-agent approach to simulate in a realistic way the traffic phenomenon at junction. We use a multi-agent coordination approach which is improved by an anticipation algorithm based on constraints processing. The resulting tool is validated by comparison between simulated flow and real flow measured at a real junction.

Implementation of Emotional Behaviours in Multi-agent System Using Fuzzy Logic and Temperamental Decision Mechanism.

Abstract: In this paper we describe our work on computational mind model for temperamental decision algorithms using Fuzzy Logic and an implementation of an emotional-behavioral multi-agent system for analysis and evaluation of different strategies based on temperamental behaviors. We describe our approach to emotional model using temperamental decision system based on theory about general types of superior nervous systems in humans and animals and we explain how we can apply Fuzzy Logic on temperamental decision system. We describe the simulation environment used in this work to test and evaluate the strategies. We have conducted a set of robotic experiments in order to test the performance of the system on its first implementation phase. The results achieved showed that the different set of temperamental characteristics influences significantly the performance of the agents and give us very positive feedback to proceed with this research and implementation in new directions.

Reputation in User Profiling for a Context-aware MultiAgent System.

Abstract: This paper aims at advancing one of the key elements in context-aware services that will make feasible the development of advanced applications of personalized agents: user profiles. These applications should be here understood as ones where mobile services, whichever they are, are provided to the final user in a distributed (agent-based) and personalized way. We examine how to provide this personalization through user profiles inferred mostly from unobtrusively monitored behavior, reducing as much as it is possible the human feedback required. Profiles will contain preferences about services, products, but also about agents representing providers and clients. This last type of preferences take the form of reputations. This part of the profile is then managed by a complex reputation model developped and tested by some of the authors in previous works. Finally, an ontological approach is applied to integrate reasoning about the three types of preferences of the user profile in the messages exchanged by agents.

An Agent Oriented Tool for Method Engineering.

Abstract: In the last years we applied the Method Engineering paradigm to the development of agent-oriented design processes. The main difficulty in our initial experiments was ensuring the support of a customised CASE tool to the new methodology. In this paper we now present a Computer-Aided Process Engineering (CAPE) tool that we developed in order to allow the design of the new process and then its instantiation. The process is executed as a workflow, the designer receives the help of an expert system (for routine works automation and syntax/semantic checks) and can model his/her system using a set of Eclipse plug-ins supporting all UML diagrams. The development of this tool started from the definition of a system metamodel obtained from the initial requirements and then instantiated using open-source and ad hoc developed components; as it could be expected relevant portions of this tool are developed using the agent-oriented paradigm.

Modeling Holonic Systems with an Organizational approach.

Abstract: Complex Systems often exhibit hierarchical structures and multiple levels of abstraction, and MultiAgent Systems, even if they have proved their adequacy to model such systems, still remains in their larger part at one or two abstraction levels. It seems uncertain in this context that MAS will be able to catch efficiently the whole complexity of such systems. Holonic Multi-Agent Systems and their intrinsic hierarchical structure appears as a natural alternative. In this paper we propose a generic framework for HMAS modeling based upon an organizational approach. The means to model a compound holon using organizations and how to conciliate in a generic way the holonic structure with problem-dependent behaviors, are detailed. Different aspects related to the holon dynamics : holon creation, members integration and self-organization are also studied.

A Multi-agent Approach for the Edge Detection in Image Processings.

Abstract: Several multi-agent approaches have been proposed to improve image processing. They use several image processing algorithms simultaneously. However, these approaches do not deal with the inherent problems encountered for the extraction from an image of primitive information like edges or regions. This implies that agents use macro results provided by image processing algorithms. Agents use macro results provided by image processing algorithms. Then, the results do not take advantage of all the interesting characteristics, such as environmental adaptability and emergent behavior capability, of agent-based systems: the combinative explosion of the possible solutions offered by this kind of systems, is highly reduced. In this paper, we propose a multi-agent system based on instinctual [5] reactive agents, which are able to detect edges. Agents locally perceive their environment, that is to say, pixels and additional environmental information. This environment is built using a Kirsch derivative and a Gradient Vector Flow. Edges detection emerges from agents interaction. Problems of partial or hidden contours are solved with the cooperation between the different agents. In the scope of this paper, we illustrate our approach through an example that shows how it can be used to detect lungs on 2D images coming from a scan device.

The Effect of Heterogeneity on Coalition Formation in Iterated Request for Proposal Scenarios.

Abstract: This paper explores a general model of economic exchange between heterogeneous agents representing firms, traders, or other socioeconomic entities, that self-organise into coalitions to face specific tasks. In particular, the work addresses coalition formation problems in which many tasks are addressed to the same population over time in an iterative fashion – giving the agents the possibility to organise themselves for specific tasks. In particular, the purpose of the paper is to describe the impact of task and skill heterogeneity in the population on the performance of the agents. Experiments are carried out for two common strategies from the economic world, namely competitive and conservative strategies. Results obtained show that competitive population outperforms conservative population, and that the heterogeneity degree has a direct effect on the advantage of the first strategy versus the second. We analyse experimental results by using a novel data mining technique called collaboration graphs.

Evolutionary Stability of Behavioural Types in the Continuous Double Auction.

Abstract: In this paper, we investigate the effectiveness of different types of bidding behaviour for trading agents in the Continuous Double Auction (CDA). Specifically, we consider behavioural types that are neutral (expected profit maximising), passive (targeting a higher profit than neutral) and aggressive (trading off profit for a better chance of transacting). For these types, we employ an evolutionary game-theoretic analysis to determine the population dynamics of agents that use them in different types of environments, including dynamic ones with market shocks. From this analysis, we find that given a symmetric demand and supply, agents are most likely to adopt neutral behaviour in static environments, while there tends to be more passive than neutral agents in dynamic ones. Furthermore, when we have asymmetric demand and supply, agents invariably adopt passive behaviour in both static and dynamic environments, though the gain in so doing is considerably smaller than in the symmetric case.

Near-optimal anytime coalition structure generation

Abstract: Forming effective coalitions is a major research challenge in the field of multi-agent systems. Central to this endeavour is the problem of determining the best set of agents that should participate in a given team. To this end, in this paper, we present a novel, anytime algorithm for coalition structure generation that is faster than previous anytime algorithms designed for this purpose. Our algorithm can generate solutions that either have a tight bound from the optimal or are optimal (depending on the objective) and works by partitioning the space in terms of a small set of elements that represent structures which contain coalitions of particular sizes. It then performs an online heuristic search that prunes the space and only considers valid and non-redundant coalition structures. We empirically show that we are able to find solutions that are, in the worst case, 99% efficient in 0.0043% of the time to find the optimal value by the state of the art dynamic programming (DP) algorithm (for 20 agents), using 66% less memory.

Role-based Service Description and Discovery

Abstract: The ever-growing number of services on the WWW provides enormous business opportunities Services can be automatically discovered and invoked, or even be dynamically composed from more simples ones Agent technology provides designers with an interaction-oriented way of designing open software systems [3] In this paper we concentrate on the problem of dynamic service discovery in multi-agent systems and present a new approach that complement existing ones by considering the types of interactions that services can be used in

Specifying and Enforcing Norms in Artificial Institutions.

Abstract: In this paper we investigate two related aspects of the formalization of open interaction systems: how to specify norms, and how to enforce them by means of sanctions. The problem of specifying the sanctions associated with the violation of norms is crucial in an open system because, given that the compliance of autonomous agents to obligations and prohibitions cannot be taken for granted, norm enforcement is necessary to constrain the possible evolutions of the system, thus obtaining a degree of predictability that makes it rational for agents to interact with the system. In our model, we introduce a construct for the definition of norms in the design of artificial institutions, expressed in terms of roles and event times, which, when certain activating events take place, is transformed into commitments of the agents playing certain roles. Norms also specify different types of sanctions associated with their violation. In the paper, we analyze the concept of sanction in detail and propose a mechanism through which sanctions can be applied.

Norm-Oriented Programming of Electronic Institutions: A Rule-based Approach.

Abstract: Norms constitute a powerful coordination mechanism among heterogeneous agents. We propose means to specify and explicitly manage the normative positions of agents (permissions, prohibitions and obligations), with which distinct deontic notions and their relationships can be captured. Our rule-based formalism includes constraints for more expressiveness and precision and allows the norm-oriented programming of electronic institutions: normative aspects are given a precise computational interpretation. Our formalism has been conceived as a machine language to which other higher-level normative languages can be mapped, allowing their execution.

An Argumentation-based Approach for Dialog Move Selection.

Abstract: Modeling different types of dialog between autonomous agents is becoming an important research issue. Several proposals exist with a clear definition of the dialog protocol, which is the set of rules governing the high level behavior of the dialog. However, things seem different with the notion of strategy. There is no consensus on the definition of a strategy and on the parameters necessary for its definition. Consequently, there are no methodology and no formal models for strategies.

Goal-Oriented Agent Patterns with the PRACTIONIST Framework.

Abstract: When developing BDI agent-based systems, some design patterns such as incompatible intentions, multiple strategies, intention decomposition, etc. would be very useful for specifying some desired agent behaviours. As BDI agent programmers, our desire would be to have a framework that natively supports such common patterns. The PRACTIONIST framework provides a goal-oriented approach for developing agent systems according to the BDI model. In this paper we first describe the goal model of PRACTIONIST agents and how they use such a model to reason about goals during their deliberation process and means-ends reasoning. Then, we show how some useful BDI agent patterns can be directly and actually implemented with our framework, which natively supports such designlevel solutions. In other words, in our framework we wanted to solve some common design problems, by providing some built-in solutions that programmers can easily adopt when developing their intentional agents.

A Utility-Based Sensing and Communication Model for a Glacial Sensor Network.

Abstract: This paper reports on the development of a utility-based mechanism for managing sensing and communication in cooperative multi-sensor networks. The specific application considered is that of GLACSWEB, a deployed system that uses battery-powered sensors to collect environmental data related to glaciers which it transmits back to a base station so that it can be made available world-wide to researchers. In this context, we first develop a sensing protocol in which each sensor locally adjusts its sensing rate based on the value of the data it believes it will observe. Then, we detail a communication protocol that finds optimal routes for relaying this data back to the base station based on the cost of communicating it (derived from the opportunity cost of using the battery power for relaying data). Finally, we empirically evaluate our protocol by examining the impact on efficiency of the network topology, the size of the network, and the degree of dynamism of the environment. In so doing, we demonstrate that the efficiency gains of our new protocol, over the currently implemented method over a 6 month period, are 470%, 250% and 300% respectively.

Creating a Comprehensive Agent-oriented Methodology.

Abstract: In an electronic pattern generator characters to be recorded are displayed on the screen of a cathode ray tube and then exposed on a recording medium mounted on a carrier, which is displacable in a plane parallel to the plane of the screen. The characters are displayed from video-data recallable from a store. The recalled video-data are transformed into a first video signal for the brightness control of the electron beam of the cathode ray tube into first deflection signals for the beam and into displacement signals for the carrier. For controlling the make-up and composing process the characters are displayed identically on a further screen of a soft copy terminal independently of the display on the screen of the cathode ray tube. By identical display it is understood, that the various type fonts and sizes of characters are not displayed in an unitary fashion, but in the original font of writing, as they will appear later in the typographical composition on the recording medium. By identical display it is further understood, that also the appropriate displacement motions of the carrier are taken into account when displaying the soft copy for control purpose. The identical display is achieved by storing the first video signal and transforming it into a second video signal for the soft copy terminal with the help of different read-in and read-out signals for the storage means and different second deflection signals for the screen of the soft copy terminal.

Self-regulating Multi-agent System for Multi-disciplinary Optimisation Process.

Abstract: This article presents a multi-agent method to tackle multidisciplinary optimisation, based on the notions of cooperation and self-regulation. It is focused on the preliminary aircraft design, which is a complex compromise. In our approach several cooperative agents collectively act to achieve a common goal, i.e. optimising a multi-objective function, even if the environment of the system (the user’s requirements) changes during the solving process. In MASCODE, one agent encapsulates one discipline and is designed individually without considering the dependencies with the others. So the computation is conceptually distributed without central control. Experimental results including efficiency comparison with the classical FSQP method are presented, and show that the adaptive behaviour of MASCODE provides new capabilities to understand and manage the complexity of the preliminary aircraft design.

OWL-S for Describing Artifacts.

Abstract: Artifacts for Multi-Agent Systems have been defined as runtime entities providing some kind of function or service that agents can fruitfully exploit to achieve their individual or social goals In order to enable agents to autonomously discover the available services, select the proper artifact(s), and use the services they need, it is necessary to describe the services offered by artifacts in terms of (i) what each service does, (ii) how to access it, and (iii) how it works In this paper we explore how OWL-S (Ontology Web Language for Services), a language specifically aimed for the description of Web Services, can be exploited also for describing the services offered by artifacts In particular, we investigate how to represent the artifact features by means of OWL-S, provide an example, and discuss some preliminary results in terms of advantages and limitations of this approach