Because of the gate effect (and perhaps others) that prevent mutated networks being resimulated if there has been no change. Although in some cases, the network will have changed, but not the fitness, so the network will show as one thing, but the timeseries plotted will be outdated.
Could do with rerunning the simulation for the fittest network after evolution is over.
Because of the gate effect (and perhaps others) that prevent mutated networks being resimulated if there has been no change. Although in some cases, the network will have changed, but not the fitness, so the network will show as one thing, but the timeseries plotted will be outdated.
Could do with rerunning the simulation for the fittest network after evolution is over.