Democratic Re nement in a Multi-Agent World
نویسنده
چکیده
This project is concerned with how a group of agents can work together to correct faults in their own performance. We believe that reenement can be transformed into a DAI problem which agents will cooperate to solve. All agents in the community would have the opportunity to contribute to this process; hence the word democratic in the project title. The ideas in this document are the culmination of a literature review of three distinct research areas: Distributed Artiicial Intelligence (DAI), Machine Learning in DAI systems and Theory Reenement. Initially we looked at the general uses of Machine Learning in distibuted systems. This led to the observation that no research exists to date on reening the knowledge in a DAI system. Therefore we decided to investigate this speciic area. Section 2 is a general overview of the reasons for studying DAI and the research to date in this area. In Section 3 the limited existing research on integrating learning into DAI systems is reviewed. Sections 3 and 4 outline the reasons why reenement is necessary in a DAI system, the current approaches to reenement of a single theory or knowledge base and why a new approach is required in order to reene knowledge in a DAI system. Finally our current research objectives and projected future work are set out in Section 5. At this stage I will introduce an example DAI system which is used to illustrate various points throughout this document. It is a predator-prey scenario of a kind that is often used in the DAI literature. A system of hunters cooperate in order to capture large prey. There are two diierent types of agents: scouts and assassins. Scouts have better perception than assassins but are physically weaker and cannot kill large prey on their own although they may attack prey smaller than themselves. Assassins nd it advantageous to cooperate with scouts in order to locate prey which is then shared among all those who participated in the hunt. The number of hunters needed to ensure a kill will depend on the species of prey. Communication is possible between all hunters. All agents may be peers or there may be a pack leader who directs the other hunters. The following is a reproduction of a slide 1
منابع مشابه
A Planning Agent on the World Wide Web
Work is described which seeks to support multi agent mixed initiative inter action between a task assignment or command agent and a planning agent Each agent maintains an agenda of outstanding tasks it is engaged in and uses a common representation of tasks plans processes and activities based on the notion that these are all constraints on behaviour Interaction between the agents uses explicit...
متن کاملA Comparison of Three Adaptive Finite Element Re nement Techniques for Incompressible Navier-Stokes Equations Using a CBS Scheme
The numerical solution of incompressible Navier-Stokes equations for ows involving complex geometries is greatly a ected by mesh resolution. In these ows, some regions may need ner mesh than others. Adaptive Mesh Re nement (AMR) techniques enable the mesh to be locally re ned, based on error distribution in the previous analysis. In this paper, three adaptive re nement methods, namely, Supercon...
متن کاملAnisotropic Patch-Based Adaptive Mesh Re nement for Finite-Volume Method
We propose an anisotropic, patch-based adaptive mesh re nement algorithm for the nite-volume method solving partial di erential equations on Cartesian and mapped grids. For large ow gradients, such as shock waves and ame fronts, that are aligned with one grid direction, anisotropic re nement can provide a similar reduction in error using much less grid cells compared to isotropic re nement. To ...
متن کاملMulti agent Planning via Mutually Constraining the Space of Behaviour Position Paper
Work is described which seeks to support multi agent mixed initiative interaction between a task assignment or command agent and a planning agent Each agent maintains an agenda of outstanding tasks it is engaged in and uses a common representation of tasks plans processes and activities based on the notion that these are all constraints on behaviour Interac tion between the agents uses explicit...
متن کاملA Multistrategy Learning Scheme for Agent Knowledge Acquisition
The problem of designing and re ning task-level strategies in an embedded multiagent setting is an important unsolved question. To address this problem, we have developed a multistrategy system that combines two learning methods: operationalization of high-level advice provided by a human and incremental re nement by a genetic algorithm. The rst method generates seed rules for ner-grained re ne...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 1994