Catálogo de publicaciones - libros

This research paper will outline the new Plug-in Oriented Agent Architecture (POAA) and describe the agents that make use of the POAA architecture. POAA agents make heavy use of functional and controller based plug-ins in order to extend the functionality and behavior of the agent. The architecture was designed to facilitate machine learning and agent mobility techniques. POAA agents are created by mounting newly created dynamic plug-in components into the static structure of the agent. The use of plug-ins will greatly improve the effectiveness of researchers, as only a single, standard architecture will exist. Researchers only need to design and develop the plug-in required for their specific agent to function as desired. This will also facilitate the comparison of various tools and methods, as only the components being reviewed need to be interchanged to measure system performance.

When a decision-maker needs to choose among a number of choices, each having a certain probability to happen, one of the traditional ways discussed in economics is to calculate the expected utility of each choice and chooses the choice with the maximum expected utility. However, most of the humans do not do so in real situations. One of the famous example is the Allais paradox. The reason why most of the people do not maximize the expected utility is that people have different attitudes towards risk in different situations and people are generally risk-averse. In this paper, we proposed a model of decision-making, which considers the risk attitude of a player, reputations of other players and the payoffs of the choices. We call this model . We find that the model of attractiveness maximization can be used to explain a situation, the Allais paradox, which violates the expected utility theory.

Combinatorial auctions can be used to reach efficient resource and task allocations in multi-agent system where the items are complementary or substitutable. Determining the winners is NP-complete. This paper presents a negotiation model which is based on the multi-agent, and gives the basic characteristics and the satisfied conditions of winner determination which is determined by the solution of Nash negotiation and indicate the rationality and fairness of this method based on the analysis about the utility increment of the bid agent and the benefit increment of the auction agent.

Web Service (WS) has been widely accepted in recent years. It is becoming the next paradigm to deploy business services on the Web. However, building WS is not an easy task due to the complexity of various business processes and different communication protocols. In this paper, we propose a methodology to efficiently develop Web Services by plugging in software agents. This agent oriented method uses software agents as building blocks of WS, and exploits the commonly available agent development tools to accelerate the whole development cycle. Software agents in this methodology not only implement the business processes, but also enrich the functions of WS. Our approach provides a solution for Web Service engineering. A case study in loan application service is presented to show the benefits and advantages of this agent oriented service engineering methodology.

The RoboCup Soccer Simulation league is brought to public attention as an intriguing application of agent technology. In this paper, we introduce argumentation mechanism into soccer agents, in order to enhance soccer agents of reactive nature with an ability of deliberation. In doing so, we take into account resource-boundedness and acceleration of argumentation for deliberate but quick decision and action. We demonstrate some advantages of arguing soccer agents with a good balance between reactivity and deliberation with experimental results.

Agent-based computing is an important research area. Agent and multi-agent system can be used to model real systems in complex and dynamic environments. However, most of them assume that there is no uncertain and fuzzy information in an agent’s mental state and environment. In this paper, we remove this unrealistic assumption and propose a new agent programming language which allows agent programs to effectively perform with fuzzy knowledge under uncertain environment, and to dynamically adapt with changes of the environment. This language presents a new and practical approach to solve fuzzy constraints in uncertain environment which consists of three components for programming agent: uncertain belief updating, goal selection and revision and uncertain practical reasoning.

In Grid systems, which have become more and more bigger and more complex, it is critical to utilize heterogeneous resources efficiently. In this paper, we designed and implemented an agent-based monitoring system, which consists of dynamically controllable agents. The structure of the monitoring system is divided into three layers to archive independence among communication protocol, message interpretation, and actual tasks. Components of each layer can be replaced with other components whenever they are required. To reduce development times for monitoring application in distributed systems, four monitoring agents were provided, which help to reuse legacy applications through user query.

In ubiquitous environment where all entities can freely connect and collaborate with each other from anywhere, the amount of accessible information is overwhelming and desired information often remains unfound. So there is a growing need to provide the personalized recommendation services for the customers in ubiquitous space. This paper suggests a UREC_P2P(U-RECom-mendation by peer-to-peer), a recommendation procedure in ubiquitous environment adopting P2P technologies combined with collaborative filtering algorithm. UREC_P2P is implemented and comparatively evaluated with a CF-based recommender system in client-server environment. The evaluation result shows that UREC_P2P has a good potential to be a preeminent and realistic solution to the recommendation problems encountered in ubiquitous environment.

In order to support the advanced concept technology demonstration of intelligence reconnaissance activities, multiple sensors simulation needs to provide enough detail to examine sensing dynamics on future battlefield. However, most current researches lack a modeling mechanism based on platform-level agent interactions and limit the implementation of agent-based simulations. In this paper, platform-level agent interactions based modeling and simulation technology is proposed to solve the problem. According to agent definition of mapping function from perception sequences to actions, the mapping from multiple sensors system’s members to respective agents is set up. Thus, the sensor entity agent model is designed. Moreover, sensor agent interactions model is presented to support multiple sensors simulation based on multi-agent interactions by using an improved Contract Net Protocol, where state transitions and augmented transition network are studied to represent the interactions. The established demonstration system proves the feasibility and efficiency of our model.

A computer virus is a program that can generate possibly evolved copies of itself when it runs on a computer utilizing the machine’s resources, and by some means each copy may be propagated to another computer in which the copy will have a chance to get executed. And we call a virus instance as a viral agent since it is autonomous during its execution by choosing what action to perform in the computer without a user’s intervention. In the paper we develop a computational model of viral agents based on the persistent Turing machine (PTM) model which is a canonical model for sequential interaction. The model reveals the most essential infection property of computer viruses well and overcomes the inherent deficiency of Turing machine (TM) virus models in expressing interaction. Then on that basis we deduce several helpful theorems about viral agents. Finally we also discuss modeling of viral agent dynamics with cellular automata (CAs) and get some useful results.