Fluctuating crosstalk as a source of deterministic noise and its effects on GA scalability
17 December 2005Sastry, K., Winward, P., Goldberg, D. E., Lima, C. F. (2006). Applications of Evolutionary Computing EvoWorkshops2006: EvoBIO, EvoCOMNET, EvoHOT, EvoIASP, EvoInteraction, EvoMUSART, EvoSTOCK. 740—751. [Full paper - PDF] [Full paper - PS].
Abstract:
-
This paper explores how fluctuating crosstalk in a deterministic fitness function introduces noise into genetic algorithms. We model fluctuating crosstalk or nonlinear interactions among building blocks via higher-order Walsh coefficients. The fluctuating crosstalk behaves like exogenous noise and can be handled by increasing the population size and run duration. This behavior holds until the strength of the crosstalk far exceeds the underlying fitness variance by a certain factor empirically observed. Our results also have implications for the relative performance of building-block-wise mutation over crossover.
Posted in Principled Efficiency Enhancement Techniques, Genetic and Evolutionary Algorithms, Conference Proceedings, Publications | Trackback | del.icio.us | Top Of Page
No comments yet