To say that memetic algorithms are like some part or some strategy that we would use is one thing, but to believe that these are in themselves solutions to the more confounding problems of contemporary AGI is to have been duped as well as confounded.
On Fri, Dec 7, 2012 at 2:07 PM, Jim Bromer <[email protected]> wrote: > Tim Tyler <[email protected]> wrote: > Powerful machine intelligence programs will likely be memetic > algorithms. > > > I think this is naive. Perhaps there is something I am missing but I > believe that you are taking the hype literally. > > A contemporary computer program is not going to be able to represent a > meme or a concept with the same degree of richness that a human being > might. I see that. However, that does not mean that a meme can be > adequately represented in the same way a superficial string in a Genetic > Algorithm can. > > And I appreciate the fact that GAs and MAs are interesting. An aspect that > I have called rational imagination (that might not be a good name) is > somewhat like recombining parts of concepts. Doing so while using > partially serendipitious acquired insight makes it much more > interesting. But as we can thereby use 'concept'-things to find reasons > and methods which are themselves concept-things to recombine other > 'concept'-things this makes the process relativistic. Yes the processes > can be limited in certain circumstances, but we also need to figure out why > something did not work or why it did work. And those are problems which > sometimes require expanding the search space. This in turn can lead to new > insights but it also obviously complex. > > To say that memetic algorithms are like some part or some strategy that we > would use is one thing, but to believe that these are in themselves > solutions to the more confounding problems of contemporary AGI is to > have duped as well as confounded. > > Jim Bromer > > > > > > > > > > On Thu, Dec 6, 2012 at 9:36 PM, Tim Tyler <[email protected]> wrote: > >> On 06/12/2012 15:50, Jim Bromer wrote: >> >> Memetic algorithms sound like more of the same. Maybe I am not getting it >> but it doesn't sound like it is going to lead to anything that less >> formalized methods haven't been able to do. It seems obvious to me that a >> memetic algorithm is not a breakthrough method that would make an AGI >> program feasible. >> >> >> Powerful machine intelligence programs will likely be memetic >> algorithms. Just as genetic >> algorithms attempted to mimic organic evolution so memetic algorithms are >> inspired by >> cultural evolution. Rather obviously, it's cultural evolution that will >> lead to machine >> intelligence. However, part of why cultural evolution reached take-off >> point is because >> humans have big brains. So, attempting to emulate the process in >> machines leads to >> a bit of a chicken-and-egg situation. >> -- >> __________ >> |im |yler http://timtyler.org/ [email protected] Remove lock to reply. >> >> *AGI* | Archives <https://www.listbox.com/member/archive/303/=now> >> <https://www.listbox.com/member/archive/rss/303/10561250-470149cf> | >> Modify<https://www.listbox.com/member/?&;>Your Subscription >> <http://www.listbox.com> >> > > ------------------------------------------- AGI Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657 Powered by Listbox: http://www.listbox.com
