Agent-Based Evolutionary Search (Adaptation, Learning, and Optimization, Volume 5)

Agent-Based Evolutionary Search (Adaptation, Learning, and Optimization, Volume 5)

Language: English

Pages: 293


Format: PDF / Kindle (mobi) / ePub

The performance of Evolutionary Algorithms can be enhanced by integrating the concept of agents. Agents and Multi-agents can bring many interesting features which are beyond the scope of traditional evolutionary process and learning.

This book presents the state-of-the art in the theory and practice of Agent Based Evolutionary Search and aims to increase the awareness on this effective technology. This includes novel frameworks, a convergence and complexity analysis, as well as real-world applications of Agent Based Evolutionary Search, a design of multi-agent architectures and a design of agent communication and learning Strategy.












Journal, vol. 1(1), pp. 7–38. Kluwer Academic Publishers, Boston (1998) [3] Ferber, J.: Multi-Agent Systems: An Introduction to Distributed Artificial Intelligence. Addison-Wesley, New York (1999) [4] Liu, J.: Autonomous Agents and Multi-Agent Systems: Explorations in Learning, Self-Organization, and Adaptive Computation. World Scientific, Singapore (2001) [5] Liu, J., Tang, Y.Y., Cao, Y.C.: An evolutionary autonomous agents approach to image feature extraction. IEEE Trans. Evol. Comput. 1(2),

well-known benchmark problems including two new problems. The experimental results confirm the improved performance of the proposed algorithm. 1 Introduction Over the last few decades, Evolutionary Algorithms (EAs) have been widely adopted to solve optimization problems due to their flexibility and adaptability to the task at hand, in combination with their robust performance and global search Abu S.S.M. Barkat Ullah School of Engineering and Information Technology (SEIT), UNSW@ADFA, Canberra

called Mission Capability Planning Problem (MCPP) [14], we propose to build a two-level hierarchical system for risk analysis. At the first level, a planning engine is used to produce options for a particular planning problem. The second layer is in charge of managing DM’s attitudes towards risk. For the first layer, the problem is initially analyzed in the context of a military capability planning process and then mathematically formulated as a multi-objective resource investment problem (a

levels of risk acceptance. We name the layer “risk tolerance layer” — RTL. The general framework is depicted in Figure 1. 4.2 Options Production Layer — OPL In this layer, a given future mission will undergo a period of pre-processing where the mission is decomposed into a series of tasks depending on the given scenarios. Basically, each task is considered a tactical event (sub-mission). Military staff usually perform the task decomposition and pre-determine the resources required by 86 L.T.

military operations planning. In: Proc. ICAPS, pp. 402–411. AAAI Press, Menlo Park (2004) 3. ADF: Defence Capability Development Manual. Australian Defence Publishing, Canberra, Australia (2006) 4. Alcaraz, J., Maroto, C.: A robust genetic algorithm for resource allocation in project scheduling. Annals of Operations Research 102(4), 83–109 (2001) 5. Azaron, A., Tavakkoli-Moghaddam, R.: Multi-objective time cost trade-off in dynamic PERT networks using an interactive approach. European Journal of

Download sample