Study of some properties of binary neutral representations

This paper analyzes some of the properties of binary neutral representations considered in the literature, using families of neutral binary representations NN
_g
(l, k) based on the mathematical formulation of error control codes. The performance of an (1 + 1)_ES modeled using Markov chains is used to verify that the representations of NN
_g
(l, k) with better performance in NK(N, K) fitness landscapes with adjacent neighborhood do not exhibit extreme values of any of the properties commonly considered in the literature of evolutionary computation.