in reply to Instrumenting a genetic algorithm.

I'll need to check with a colleague when they come in later today, but I thought that best practice in stopping genetic algorithms was still heuristically based. At least it used to be that they have a strong tendency toward overfitting when you only watch the cost function.

#11929 First ask yourself `How would I do this without a computer?' Then have the computer do it the same way.

Replies are listed 'Best First'.
Re^2: Instrumenting a genetic algorithm.
by BrowserUk (Patriarch) on Aug 19, 2016 at 16:32 UTC
    I thought that best practice in stopping genetic algorithms was still heuristically based

    The GA is heuristic based, it determines whether any given set of parameters results in success or failure; but there are a huge set of combinations of parameters to the GA algorithm. The 95% is a limit that decides which sets of parameters are worthy of further, exhaustive examination.

    With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday'
    Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
    "Science is about questioning the status quo. Questioning authority". I knew I was on the right track :)
    In the absence of evidence, opinion is indistinguishable from prejudice.
[reply]

In Section Seekers of Perl Wisdom