40Re: [genetic_programming] Beowulf Clusters
- Oct 1, 2001
> Our plan is to have one computer hold the population, and farm outWell, this has, as I understand it, traditionally been avoided because
> evaluations to all computers. So, the faster computers will simply get
> a higher percentage of the population to evaluate. This has a lot of
> nice properties, including that we can repeat runs by simply using the
> same seed, assuming evaluation is deterministic.
1. it scales worse than multiple separate demes, only exchanging about 5%
of the population (I guess that's 5% each way=10%) cuts down on
2. separate demes have been suggested to have beneficial properties over
one huge population.
I not really in any position to comment on either claim, but it's
something to think about.
> I think the "only exchange info if within 10 generations" thing isYes, this is true, especially since one run might pass the boundry over
> particularly dangerous; in one run, populations may be completely
> separate from almost the start, whereas in another, they'll be sharing
> info most of the way through. That's a very strange dynamic.
to the next run, and never ever meet the rest of the demes again (they'd
be on earlier runs all the time). Restarting all runs at the same time
would take a lot of weird synchronization.
- << Previous post in topic Next post in topic >>