Learning by Exploration and Cooperation in Dynamic Agencies
نویسندگان
چکیده
The building of cooperative multi-agent systems is an issue of growing importance in the field of Distributed Artificial Intelligence. We introduce a type of agency, called dynamic agency, that overcomes the critical problem of having agents that have both several specific operation functions, and one uniform cooperation ability. A dynamic agency is also suited to give flexibility with respect to classes of problems that can be addressed by the agency. Dynamic agencies are built from intelligent agents only providing operations (Fixed Intelligent Agents) that are visited by an intelligent agent carrying cooperation (Mobile Intelligent Agent), which replicates itself together with each fixed agent, in order to complete the agency based on a biagent fixed-mobile structure. The aim of the paper is to illustrate two relevant learning aspects that are of basic importance in constructing dynamic agencies. The first one is the exploration of fixed intelligent agents performed by a visiting mobile agent, namely the sequence of extraction of knowledge from fixed intelligent agents and organization of cooperation. The second learning aspect is related to the construction, called cooperation, of the strategic planner of the dynamic agency, that is devoted to decompose an agency problem in simpler agent sub-problems tackled by tactical planners.
منابع مشابه
Use of Evidence-Informed Deliberative Processes – Learning by Doing; Comment on “Use of Evidence-informed Deliberative Processes by Health Technology Assessment Agencies Around the Globe”
The article by Oortwijn, Jansen, and Baltussen (OJB) is much more important than it appears because, in the absence of any good general theory of “evidence-informed deliberative processes” (EDP) and limited evidence of how they might be shaped and work in institutionalising health technology assessment (HTA), the best approach seems to be to accumulate the experience of...
متن کاملExploration of Arak Medical Students’ Experiences on Effective Factors in Active Learning: A Qualitative Research
Introduction:: Medical students should use active learning to improve their daily duties and medical services. The goal of this study is exploring medical students’ experiences on effective factors in active learning. Methods: This qualitative study was conducted through content Analysis method in Arak University of Medical Sciences. Data were collected via interviews. The study started with p...
متن کاملThe Identification and Exploration of the Culture of Elementary School Teachers in Lorestan Province
The Identification and Exploration of the Culture of Elementary School Teachers in Lorestan Province S. Roomaani* G.R. Haji Hosseinnejaad, Ph.D.** A. Hosseinikhaah, Ph.D.*** N. Faazeli, Ph.D.**** In centralized systems of education, the incongruence between the national and local cultures can be problematic. Hence, the identification of the local culture and exploration of its congru...
متن کاملDynamic Obstacle Avoidance by Distributed Algorithm based on Reinforcement Learning (RESEARCH NOTE)
In this paper we focus on the application of reinforcement learning to obstacle avoidance in dynamic Environments in wireless sensor networks. A distributed algorithm based on reinforcement learning is developed for sensor networks to guide mobile robot through the dynamic obstacles. The sensor network models the danger of the area under coverage as obstacles, and has the property of adoption o...
متن کاملTeaching-Learning approach in complexity paradigm
"Teaching-Learning Approach" is a model of interaction between teachers and students in an educational environment and one of the main components of the educational system. This model can be organized and designed on the basis of various opinions and ideas, including philosophical or scientific theories. This research aims to design and explain teaching-learning approach based on the complexity...
متن کامل