Marek Kisiel-Dorohinicki

Agent-Oriented Model of Simulated Evolution

The paper deals with a specific class of multi-agent systems, in principle similar to evolutionary algorithms, but utilising a more complex, since decentralised, model of evolution. The proposed layered architecture uses the notion of a profile that models strategies and goals of an agent with respect to some aspect of its operation. The paper presents main ideas of the architecture illustrated by a concrete realisation that is an evolutionary multi-agent system solving a generic optimisation problem.

Evolutionary Multi-Agent Systems

The aim of this paper is to give a survey on the development and applications of evolutionary multi-agent systems (EMAS). The paper starts with a general introduction describing the background, structure and behaviour of EMAS. EMAS application to solving global optimisation problems is presented in the next section along with its modification targeted at lowering the computation costs by early removing certain agents based on immunological inspirations. Subsequent sections deal with the elitist variant of EMAS aimed at solving multi-criteria optimisation problems, and the co-evolutionary one aimed at solving multi-modal optimisation problems. Each variation of EMAS is illustrated with selected experimental results.

Security, energy, and performance-aware resource allocation mechanisms for computational grids

Distributed Cyber Physical Systems (DCPSs) are networks of computing systems that utilize information from their physical surroundings to provide important services, such as smart health, energy efficient grid and cloud computing, and smart security-aware grids. Ensuring the energy efficiency, thermal safety, and long term uninterrupted computing operation increases the scalability and sustainability of these infrastructures. Achieving this goal often requires researchers to harness an understanding of the interactions between the computing equipment and its physical surroundings. Modeling these interactions can be computationally challenging with the resources on hand and the operating requirements of such systems. In this paper, we define the independent batch scheduling in Computational Grid (CG) as a three-objective global optimization problem with makespan, flowtime and energy consumption as the main scheduling criteria minimized according to different security constraints. We use the Dynamic Voltage Scaling (DVS) methodology for reducing the cumulative power energy utilized by the system resources. We develop six genetic-based single- and multi-population meta-heuristics for solving the considered optimization problem. The effectiveness of these algorithms has been empirically justified in two different grid architectural scenarios in static and dynamic modes.

Agent-based evolutionary multiobjective optimisation

This work presents a new evolutionary approach to searching for a global solution (in the Pareto sense) to a multiobjective optimisation problem. The novelty of the method proposed consists in the application of an evolutionary multi-agent system (EMAS) instead of classical evolutionary algorithms. Decentralisation of the evolution process in EMAS allows for intensive exploration of the search space, and the introduced mechanism of crowd allows for effective approximation of the whole Pareto frontier. In the paper the technique is described as well as preliminary experimental results are reported.

Computing agents for decision support systems

In decision support systems, it is essential to get a candidate solution fast, even if it means resorting to an approximation. This constraint introduces a scalability requirement with regard to the kind of heuristics which can be used in such systems. As execution time is bounded, these algorithms need to give better results and scale up with additional computing resources instead of additional time. In this paper, we show how multi-agent systems can fulfil these requirements. We recall as an example the concept of Evolutionary Multi-Agent Systems, which combines evolutionary and agent computing paradigms. We describe several possible implementations and present experimental results demonstrating how additional resources improve the efficacy of such systems. Applicability of agent-oriented metaheuristics to decision support systems.Functional-programming based prototypes of agent-based computing systems.Experiments regarding scalability and performance of the implemented systems.

Agent Populations as Computational Intelligence

The paper deals with agent-based architectures of hybrid soft-computing systems, which should exhibit intelligent behaviour at a population level. Three levels of complexity of such systems are distinguished together with their potential advantages. The considerations are illustrated by prototypical realisations of evolutionary multi-agent systems dedicated to multiobjective optimisation, data classification, and time-series prediction.

Agent-Based Evolutionary and Immunological Optimization

An immunological selection mechanism for evolutionary multi-agent systems is discussed in the paper. It allows for reducing the number of fitness assignments required to get the solution of comparable quality as the classical resource-based selection in EMAS. Experimental studies aim at comparing the performance of immune-inspired selection, with resource-based one, and also with classical parallel evolutionary algorithms, based on typical multi-modal optimization benchmarks.

Immunological Selection Mechanism in Agent-Based Evolutionary Computation

Artificial immune systems turned out to be interesting technique introduced into the area of soft-computing. In the paper the idea of an immunological selection mechanism in the agent-based evolutionary computation is presented. General considerations are illustrated by the particular system dedicated to function optimization. Selected experimental results conclude the work.

Agent-Based Model and Computing Environment Facilitating the Development of Distributed Computational Intelligence Systems

In the paper a simple formalism is proposed to describe the hierarchy of multi-agent systems, which is particularly suitable for the design of a certain class of distributed computational intelligence systems. The mapping between the formalism and the existing computing environment AgE is also sketched out.

Agent-Based Models and Platforms for Parallel Evolutionary Algorithms

The goal of the paper is to provide an overview of classical and agent-based models of parallel evolutionary algorithms. Agent approach reveals possibilities of unification of various models and thus allows for the development of platforms supporting the implementation of different PEA variants. Design considerations based on AgWorld and Ant.NET projects conclude the paper.