. Adaptation in open, multi-agent information gathering systems is important for several reasons. These reasons include the inability to accurately predict future problem-solving workloads, future changes in existing information requests, future failures and additions of agents and data supply resources, and other future task environment characteristic changes that require system reorganization. We have developed a multi-agent distributed system infrastructure, Retsina (REusable Task Structure-based Intelligent Network Agents) that handles adaptation in an open Internet environment. Adaptation occurs both at the individual agent level as well as at the overall agent organization level. The Retsina system has three types of agents. Interface agents interact with the user receiving user specifications and delivering results. They acquire, model, and utilize user preferences to guide system coordination in support of the user's tasks. Task agents help users perform tasks by formulating p...

Complex, real-world domains require a rethinking of traditional approaches to AI planning. Planning and executing the resulting plans in a dynamic environment requires a continual approachinwhich planning and execution are interleaved, there may be uncertaintyin the current and projected world state, and replanning may be required when the situation changes or planned actions fail. Furthermore, complex planning and execution problems may require multiple computational agents and human planners to collaborate on a solution. In this article, we describe a new paradigm for planning in complex, dynamic environments, whichweterm distributed,continual planning (DCP). We argue that developing DCP systems will be necessary in order for planning applications to be successful in these environments. We give a historical overview of research leading up to the current state of the art in DCP, and describe research in distributed and continual planning. The increasing emphasis on r...

Much recent planning research has focused on two related issues. First, there has been a strong interest in information-gathering (or "sensing", or "knowledge-producing ") actions. Second, has been an investigation of plans with sophisticated control structures, such as conditional branches and loops. But the combination of these two lines of research poses a representational problem: plans with information-gathering actions that can be executed more than once can have complex information-flow and control-flow relationships. In this paper, we present a framework for the representation and execution of hierarchical plans with information producing actions, conditional branches, periodic actions, and loops. Our framework subsumes several techniques found in the recent literature.

Amalthaea is an evolving, multiagent ecosystem for personalized filtering, discovery and monitoring of information sites. Amalthaea's primary application domain is the World-Wide-Web and its main purpose is to assist its users in finding interesting information. Two different categories of agents are introduced in the system: filtering agents that model and monitor the interests of the user and discovery agents that model the information sources. A market-like ecosystem where the agents evolve, compete and collaborate is presented: agents that are usefull to the user or other agents reproduce while low-performing agents are destroyed. Results from various experiments with different system configurations and varying ratios of user interests vs agents in the system are presented. Finally issues like fine-tuning the initial parameters of the system and establishing and maintaining equilibria in the ecosystem are discussed. Keywords: Agents, Evolution, Information Filtering, World-Wide-Web...

The World Wide Web has become an invaluable information resource but the explosion of available information has made web search a time consuming and complex process. The large number of information sources and their different levels of accessibility, reliability and associated costs present a complex information gathering control problem. This paper describes the rationale, architecture, and implementation of a next generation information gathering system – a system that integrates several areas of Artificial Intelligence research under a single umbrella. Our solution to the information explosion is an information gathering agent, BIG, that plans to gather information to support a decision process, reasons about the resource trade-offs of different possible gathering approaches, extracts information from both unstructured and structured documents, and uses the extracted information to refine its search and processing activities.

Inexperienced users typically obtain one of three possible outcomes when they search for online information: they are buried under an information avalanche, they are unable to locate any useful information at all, or they find what they need in roughly the amount they need. Unfortunately, the latter outcome is the most rare. Unfamiliarity with search tactics creates difficulties for many users of online retrieval systems. When faced with poor results, even experienced searchers may use vocabulary incorrectly and often fail to reformulate their queries. Far from being the answer to everyone's information dreams, distributed sources of online information, i.e., the World Wide Web (WWW), compound the problem and may often turn into an information nightmare. To address this problem, intelligent online search assistants, or agents, are being developed for information retrieval applications. There are many approaches, both theoretical and implemented, to using intelligent software agents for...

One aspect of multi-agent systems (MAS) that has been only partially studied is their role in software engineering, and in particular their merit as a software architecture style. As we demonstrate, multi-agent systems developed to date have several common architectural characteristics, even though differences in their design and implementation result in variations in their strengths and weaknesses. A large portion of the research in the design and implementation of MAS addresses questions such as: given a computational problem---can one build a MAS to solve it? What should be the properties of this MAS given the problem? Having developed a MAS, what is the class of problems that this MAS, either as developed or with slight modifications, can solve? MAS research has provided several answers to the questions above. However, more fundamental questions were left un-answered: given a computational problem---is a MAS an appropriate solution? If it is, what type of MAS should be preferred? ...

Scheduling complex problem solving tasks, where tasks are interrelated and there are multiple different ways to go about achieving a particular task, is an imprecise science and the justification for this lies soundly in the combinatorics of the scheduling problem. Intractable problems require approximate solutions. We have developed a new domain-independent approach to task scheduling called Design-to-Criteria that controls the combinatorics via a satisficing methodology and custom designs schedules to meet a particular client's goal criteria. In Design-to-Criteria, criteria directed focusing, approximation, and heuristics, in conjunction with soft goal criteria are used to make the scheduling problem tractable. We describe the interesting facets of the Design-to-Criteria approach and give examples of its power at reducing the complexity of the scheduling task while designing custom satisficing schedules.

) Mike Williamson and Keith Decker and Katia Sycara The Robotics Institute, Carnegie-Mellon University 5000 Forbes Ave., Pittsburgh, PA 15213 (mikew,decker,sycara)@cs.cmu.edu April 15, 1996 1 Introduction An agent that carries out plans in a realistic domain faces many difficulties. Particular difficulties arise, though, when an agent participating in a cooperative, multi-agent environment is executing decision-theoretic plans. This paper describes these specific difficulties, and discusses potential avenues to their solution. By a "cooperative, multi-agent environment," we mean a system in which agents interact to collectively solve problems. In such an environment, many of the "actions" that an agent can perform are in fact speech acts, which induce some other agent to act on the first agent's behalf. In some cases, the use of speech acts can simplify an agent's planning process by off-loading some of the planning effort. For example, if you want an airplane ticket it's much easier...

We describe Amalthaea, a system that utilizes AI methods for generating and maintaining user profiles. The profile builder we introduce uses completely distributed learning and representation mechanisms in the form of an evolving multiagent system; it observes the current and past interaction of the user with the application as well as direct user feedback to update its state and better fulfill the interests of the user. Amalthaea provides to its users personalized filtering and discovery of information; it' s primary application domain is the World-WideWeb and its main purpose is to assist users in finding interesting information. Keywords: user modeling, multiagent system, info filtering 1. Introduction The exponential increase of computer systems that are interconnected in on-line networks has resulted in a corresponding exponential increase in the amount of information available on-line. As it is becoming more and more difficult for users to cope with such amounts of information,...