Web Intelligence and Agent Systems: An International Journal - Volume 3, issue 1

Authors: Lin, Shou-de | Knoblock, Craig A.

Article Type: Research Article

Abstract: With the rapid growth of the World Wide Web, there is growing interest in developing web agents that interact with online services to acquire information. However, finding the online services perfectly suited for a given task is not always feasible. First, the agents might not be given sufficient information to fill in the required input fields for querying an online service. Second, the online service might generate only partial information. Third, the agents might need to …know the information about B by some input set A, but they can only find the online services that generate A from B. Fourth, most of the online services do not tolerate errors in the inputs, thus even a minor typo in the input field can hinder them from generating any meaningful results. This paper proposes SERGEANT, a framework for building flexible web agents that handle these imperfect situations. In this framework we exploit an information retrieval (IR) system as a general discovery tool to assist finding and pruning information. To demonstrate SERGEANT, we implemented two web agents: the Internet inverse geocoder and the address lookup module. Our experiments show that these agents are capable of generating high-quality results under imperfect situations. Show more

Citation: Web Intelligence and Agent Systems: An international journal, vol. 3, no. 1, pp. 1-15, 2005

Authors: Könönen, Ville

Article Type: Research Article

Abstract: A gradient-based method for both symmetric and asymmetric multiagent reinforcement learning is introduced in this paper. Symmetric multiagent reinforcement learning addresses the problem with agents involved in the learning task having equal information states. Respectively, in asymmetric multiagent reinforcement learning, the information states are not equal, i.e. some agents (leaders) try to encourage agents with less information (followers) to select actions that lead to improved overall utility values for the leaders. In …both cases, there are a huge number of parameters to learn and we thus need to use some parametric function approximation methods to represent the value functions of the agents. The method proposed in this paper is based on the VAPS framework that is extended to utilize the theory of Markov games, which is a natural basis of multiagent reinforcement learning. Show more

Keywords: multiagent reinforcement learning, Markov games, Nash equilibrium, Stackelberg equilibrium, value function approximation

Citation: Web Intelligence and Agent Systems: An international journal, vol. 3, no. 1, pp. 17-30, 2005

Authors: Wickramasinghe, L.K. | Alahakoon, L.D.

Article Type: Research Article

Abstract: This paper presents a multi agent system which provides a method for cooperation and coordination among agents using collective reasoning. There are two types of agents in the system: distributed agents and a central administrator agent. The distributed agents provide local knowledge extraction tasks and communicate it to the central administrator agent. The central administrator agent builds a global picture about the problem domain using the local knowledge of each agent. It uses the Growing Self …Organizing Map (GSOM), which is a dynamic version of the Self Organizing Map (SOM) for analyzing the global knowledge. The system differs from similar systems which employ central administrator functionality as the central administrator agent in the proposed system is capable of autonomously refining the task model of the distributed agents and dynamically determining the efficient number of agents required in the system. Show more

Keywords: knowledge discovery, knowledge sharing, self-organization, unsupervised learning, hierarchical clustering, data mining, distributed agents

Citation: Web Intelligence and Agent Systems: An international journal, vol. 3, no. 1, pp. 31-47, 2005

Authors: Holm Sørensen, Mikkel

Article Type: Research Article

Abstract: Facing great challenges in creating better assistive technology with ever more complex and pervasive IT systems we need a general change in our understanding of designing IT. Ways to improve the performance of IT without drowning in complexity or loosing overall normative power are of prime importance. This paper presents biomimetic design of Ambient Intelligence as a promising candidate for the task. It is shown how better assistive functionality goes hand in hand with improved infrastructure …flexibility and robustness by mimicking natural complex systems. On the basis of a presentation of organizational and developmental issues addressed by modern biological models the paper suggest preliminary design constraints for Ambient Intelligence ecologies. Show more

Citation: Web Intelligence and Agent Systems: An international journal, vol. 3, no. 1, pp. 49-61, 2005

Article Type: Other

Citation: Web Intelligence and Agent Systems: An international journal, vol. 3, no. 1, pp. 63-64, 2005

Article Type: Other

Citation: Web Intelligence and Agent Systems: An international journal, vol. 3, no. 1, pp. 65-65, 2005