CISUC

How Learning Improves the Performance of Evolutionary Agents: a Case Study with an Information Retrieval System for a Distributed Environment

Authors

Abstract

The goal of this work is to study how individual learning helps adaptive agents, living in an Artificial Life world, to develop correct behaviors when seeking for information in distributed and dynamic environments. We developed a prototype of an information retrieval system where a population of adaptive organisms browses through the environment, trying to find relevant information for a specific query. Two learning procedures are proposed and their effect is studied in different situations. Experimental results show that changes induced by learning provide an important advantage.

Cited by

Year 2007 : 2 citations

 D.Curran, C.O'Riordan, H.Sorensen An analysis of the effects of Lifetime learning on population fitness and diversity in an NK fitness landscape ECAL Lisbon, September, 2007.

 D Curran, C O'Riordan. The effects of cultural learning in populations of neural networks. Artificial life, 2007 - MIT Press

Year 2006 : 4 citations

 D. Curran and C. O'Riordan, Increasing Population Diversity Through Cultural Learning, Adaptive Behavior, 14 (4).

 D. Curran, C.O'Riordan Examining the Effects of Cultural Learning on Fitness and Diversity in Populations of Neural Networks in a Dynamic Environments 17th Irish Artificial Intelligence and Cognitive Science Conference (AICS 2006), 2006.

 D. Curran and C. O'Riordan, Sequential Task Problem Solving using Cultural Learning in Populations of Neural Networks, Artificial Life, 13 (1).

 D. Curran (2006). An Empirical Analysis of Cultural Learning: Examining Fitness, Diversity and Changing Environments in Populations of Game-Playing Network Agents. Ph.D. Thesis, National University of Ireland, Galway, Ireland.

Year 2004 : 2 citations

 D. Curran and C.O'Riordan , Evolving Agents to play connect-four using cultural learning Irish Artificial Intelligence and Cognitive Science Conference.

 D.Curran, C.O'Riordan Evolving Connect-Four Playing Neural Networks Using Cultural Learning Technicl Report of Dept. of Information Retrieval, NUI, Galway.

Year 2003 : 1 citations

 D. Mirikitani, I. Kushchu (2003). E. Coli Search: Self Replicating Agents for Web Based Information Retrieval. In Proceedings of the 4th International Conference on Intelligent Data Engineering and Automated Learning (IDEAL-2003), pp. 622-629, LNCS 2690, Springer.

Year 2002 : 1 citations

 Bauer, T. e Leake, D. (2002). Calvin: A Multi-Agent Personal Information Retrieval System. In Agent Oriented Information Systems 2002: Proceedings of the 4th International Bi-Conference Workshop (AOIS-2002), CEUR WS Proceedings.