Loading ...
Sorry, an error occurred while loading the content.

Re: [GP] evolving the fitness function

Expand Messages
  • Pete Martin
    Una-May, ... I assume you mean mechanised or automatic evolution of the fitness function? I guess most practitioners evolve fitness functions manually (i.e..
    Message 1 of 2 , Apr 1 7:31 AM
    • 0 Attachment
      Una-May,

      > I am looking for all references to work that involves evolving a fitness
      > function for genetic programming. Could you please let me know of work in
      > this vein?

      I assume you mean mechanised or automatic evolution of the fitness function?
      I guess most practitioners evolve fitness functions manually (i.e..
      debug/develop) to meet their needs (though it is seldom if ever mentioned),
      so you could throw that one in for starters.

      Just one question though. If one is to evolve a fitness function using
      anything resembling a standard evolutionary technique, wouldn't one have to
      devise a fitness measure for the fitness function being evolved? This sounds
      like it could lead one in ever decreasing circles.

      Pete
    • Una-May O'Reilly
      Pete, I m in general interested in learned fitness functions. No, I don t mean the process of developing them by hand. Let s see from the replies how people
      Message 2 of 2 , Apr 1 7:48 AM
      • 0 Attachment
        Pete,

        I'm in general interested in learned fitness functions. No, I don't mean
        the process of developing them by hand.

        Let's see from the replies how people have tackled the idea and issues.

        Una-May

        On Tue, 1 Apr 2003, Pete Martin wrote:

        > Una-May,
        >
        > > I am looking for all references to work that involves evolving a fitness
        > > function for genetic programming. Could you please let me know of work in
        > > this vein?
        >
        > I assume you mean mechanised or automatic evolution of the fitness function?
        > I guess most practitioners evolve fitness functions manually (i.e..
        > debug/develop) to meet their needs (though it is seldom if ever mentioned),
        > so you could throw that one in for starters.
        >
        > Just one question though. If one is to evolve a fitness function using
        > anything resembling a standard evolutionary technique, wouldn't one have to
        > devise a fitness measure for the fitness function being evolved? This sounds
        > like it could lead one in ever decreasing circles.
        >
        > Pete
        >
        >
        >
        > To unsubscribe from this group, send an email to:
        > genetic_programming-unsubscribe@yahoogroups.com
        >
        >
        >
        > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
        >
        >
      Your message has been successfully submitted and would be delivered to recipients shortly.