Higher-order linkage learning in the ECGA

  • Authors:
  • David Iclanzan

  • Affiliations:
  • Babes-Bolyai University, Cluj-Napoca, Romania

  • Venue:
  • Proceedings of the 14th annual conference on Genetic and evolutionary computation
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper, we present a higher-order dependency identification algorithm for binary variables that employs a novel metric to detect linkages. The proposed method applies an entropy distillation operation over bounded groups of variables. Lack of increase in randomness signals an underlaying statistical dependence between the inputs. We incorporate the higher order linkage learning in the Extended Compact Genetic Algorithm (ECGA). Experiments show that the extended linkage learning enables the ECGA to correctly model and solve problems with bounded-order building-blocks that do not contain pairwise dependencies.