When implementing NEAT I'm having some issues with the speciation distance/delta function, specifically the term N (number of genes in biggest genome).
Won't term $N$ in $δ=c1*E/N+c2*D/N+c3*W$ just out-scale $E$ and $D$, and reduce their impact on the function as the genomes grow?
The paper(3.3) recommends limiting N = 1 when genomes are small, but what happens when they grow? The impact of E and D tapers off.
For example two genomes with no similarities (silly example, but just for demonstration):
#1: 67 connections
#2: 40 connections
No similar innovation numbers, so $W$ will be 0.
$E$ = (difference in genes count) 67-40 = 27
$D$ = (total count of all genes found in one but not in other, minus excess genes) 67+40-27 = 80
$N$ = (Size of biggest genome) = 67
$c1$ = $c2$ = 1.0
$δ=1.0*27/67+1.0*80/67+0 = 1.5970$
This is pretty low compared to the default limit of $3$ that is used in the NEAT paper(4.1) examples, and by some other examples online.
Am I computing this incorrectly?
