Artiicial Autonomous Agents and Learning Systems
نویسنده
چکیده
These are the texts of two lectures delivered on the summer school for advanced information processing in Varna, Bulgaria, which was held from the 26th until the 30th of august 1994. The summer school was part of a bigger Tempusexchange programme of the european community in which Bulgarian universities and universities of the Netherlands, Greece, Belgium and the United Kingdom took part. I was a representative of the arti cial intelligence laboratory of the free university of Brussels at the time. I was working there on a project involving learning systems on arti cial autonomous agents. The Tempus exchange would not have been possible without the aid of Dr. Walter van de Velde (free university of Brussels) Dr. Peter Braspenning (university of Limburg, the Netherlands) and Dr. Vassil Vassilev (new Bulgarian university).
منابع مشابه
Artiicial Agents and Logic Programming
Artiicial agents represent a new paradigm in software engineering and Artiicial Intelligence. As complex software-controlled systems they are capable of exible autonomous behavior in dynamic and unpredictable environments. Over the past few years, researchers in computer science have begun to recognise that the technology of artiicial agents provides the key to solving many problems in distribu...
متن کاملModular Adaptivity and Behavior Based Control
Learning in animals is modular (though not encapsulated), and there is growing interest in modular, non-homogeneous models of learning within the neural networks community. Behavior-based artiicial intelligence, with its emphasis on modular components, seems an ideal platform for implementing such systems in autonomous agents. However, historically this community has shunned representation issu...
متن کاملAn Artificial Life Approach for the Synthesis of Autonomous Agents
This paper describes an evolutionary process producing dy-namical neural networks used as \brains" for autonomous agents. The main concepts used : genetic algorithms, morphogenesis process, artiicial neural networks and artiicial metabolism, illustrate our conviction that some fundamental principles of nature may help to design processes from which emerge artiicial autonomous agents. The evolut...
متن کاملStructure Adaptation in Arti cial
There is a growing evidence that the human brain follows an environmentally-guided neural circuit building that increases its learning exibility. Similarly, it has been shown that artiicial neural networks with dynamic topologies attempt to overcome the problem of determining the appropriate topology to optimally solve a given application. This paper presents a modular structure-adaptable artii...
متن کاملAgent Cooperation: Distributed Control Applications
An increasing number of natural and artiicial systems are found to be operated by autonomous agents that achieve performance comparable to those obtained under centralized control. These agents have partial knowledge of the system, make their own decisions, and communicate locally with nearby agents. They are said to operate in autonomous local mode (ALM). The overall performance can be improve...
متن کامل