[Question] Default setElitismReplacement value
Opened this issue · 3 comments
Any thoughts on why the default number for number of individuals to be replaced is set so "low" at 1 individual (out of a default population of 80?). Wouldn't a higher default number lead to improved performance for most cases "out-of-the-box"?
There are a lot of tradeoffs in this, for instance, if you increase the number of individuals that are going to go to the next population you may also reduce the diversity and increase the convergence into a local minima, etc.
There are a lot of tradeoffs in this, for instance, if you increase the
number of individuals that are going to go to the next population you may
also reduce the diversity and increase the convergence into a local minima,
etc.
On Fri, Jul 24, 2015 at 3:44 PM, ho1 notifications@github.com wrote:
Any thoughts on why the default number for number of individuals to be
replaced is set so "low" at 1 individual (out of a default population of
80?). Wouldn't a higher default number lead to improved performance for
most cases "out-of-the-box"?—
Reply to this email directly or view it on GitHub
#72.
Blog http://blog.christianperone.com | Github https://github.com/perone
| Twitter https://twitter.com/tarantulae
"Forgive, O Lord, my little jokes on Thee, and I'll forgive Thy great big
joke on me."
That is a good point, I didn't think about that. Some additional good thoughts around this topic: http://stackoverflow.com/questions/14622342/elitism-in-ga-should-i-let-the-elites-be-selected-as-parents.