Evolutionary Multi-Agent Systems is a computational paradigm proposed by Prof. Krzysztof Cetnarowicz and developed in the Intelligent Information Systems Group, which is led by Prof. Edward Nawarecki. IISG group belongs to Department of Computer Science which is a part of the AGH University of Science and Technology, one of the most famous universities located in the beautiful city of Krakow (Cracow), former capital of Poland.
Evolutionary algorithms are universal technique for general optimization, yet the selection of the proper evolutionary algorithm for the given task is an open problem. One of the reasons for that is the weakness of the theory of evolutionary algorithms: ,,We know that they work, but we do not know why'' (Back Hammel Schwefel). What is more, the model of evolution followed by most EAs (with noticeable exceptions) is much simplified and lacks many important features observed in organic evolutions. The idea of decentralised evolutionary computation, realized as an evolutionary multi-agent system may help to avoid some of the shortcomings of the model of evolutiona employed in classical EC techniques.
Among lots of issues related to agent technology one can find also evolutionary computation present in multi-agent systems (MAS). In most such cases an evolutionary algorithm is used by an agent (see graph) to aid realization of some its tasks (e.g. connected with learning or reasoning) or to support coordination of some group (team) activity. Yet it seems that interesting results may be achieved applying some model of evolution in MAS at a population level (see graph) i.e. among agents. In this case, genetic operators together with selection/reproduciton mechanisms search for (near) optimal configuration of the agents constituting a particular population or the whole system. Such a new class of adaptive multi-agent systems where evolutionary processes help to accomplish population level goals is called evolutionary multi-agent systems (EMAS).
Generally speaking, EMAS idea is based on merging classical evolutionary computation with agent-based paradigm. You can read more about EMAS in theoretical introduction. There exist some more improvements, that may be added to basic EMAS algorithm, such as niching and speciation techniques, immunological algorithms and elitist evolution.
EMAS may be applied to solving any optimization problem, particularly it was successfully applied to the problem of optimization of neural network architecture, multi-criteria optimization, multi-modal function optimization.
There are several software platforms that are developed by our researchers and their students and may be used to implement systems based on the EMAS idea. The code is available on demand, but most of the documentation is usually in Polish:
There are several papers that may be referenced in order to familiarize with EMAS idea and its applications. You can also check full bibliography of our researchers maintained by AGH-UST.
If you want any further information, feel free to contact any of the persons that are constanly involved in researching EMAS:
