University of Surrey

Test tubes in the lab Research in the ATI Dance Research

A unified framework for symbiosis of evolutionary mechanisms with application to water clusters potential model design

Le, MN, Ong, YS, Jin, Y and Sendhoff, B (2012) A unified framework for symbiosis of evolutionary mechanisms with application to water clusters potential model design IEEE Computational Intelligence Magazine, 7 (1). pp. 20-35.

[img] Text
CIM2012.pdf
Restricted to Repository staff only
Available under License : See the attached licence file.

Download (1MB)
[img] Text (licence)
SRI_deposit_agreement.pdf
Restricted to Repository staff only

Download (33kB)

Abstract

This article presents a theoretic model for facilitating the emergence of productive search profiles transpiring from the symbiosis of gene (stochastic variation) and meme (lifetime learning) working in synergy. The evolvability measure of the symbiotic search profiles for each individual is quantified by means of statistical learning on distinct sample vectors encountered along the search. The most productive search profile inferred for an individual, as defined by evolvability measure, is subsequently used to work on it, leading to the self-configuration of solvers that acclimatizes to suit the given problem of interest. Empirical studies on representative problems are presented to reflect the characteristics of symbiotic evolution. Assessment made against several recent state-of-the-art evolutionary and adaptive search algorithms highlighted the efficacy of the theoretic formalism of evolutionary mechanisms in symbiosis for autonomic search. As the design of computationally cheap advanced empirical water models for the understanding of enigmatic properties of water remains an important and unsolved problem, the article presents an illustration of symbiotic evolution for the design of (H2O)n or water clusters potential model.

Item Type: Article
Authors :
NameEmailORCID
Le, MNUNSPECIFIEDUNSPECIFIED
Ong, YSUNSPECIFIEDUNSPECIFIED
Jin, YUNSPECIFIEDUNSPECIFIED
Sendhoff, BUNSPECIFIEDUNSPECIFIED
Date : 2012
Identification Number : 10.1109/MCI.2011.2176995
Depositing User : Symplectic Elements
Date Deposited : 28 Mar 2017 14:42
Last Modified : 31 Oct 2017 14:34
URI: http://epubs.surrey.ac.uk/id/eprint/532126

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year


Information about this web site

© The University of Surrey, Guildford, Surrey, GU2 7XH, United Kingdom.
+44 (0)1483 300800