home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!walter!att-out!rutgers!uwm.edu!zaphod.mps.ohio-state.edu!rpi!newsserver.pixel.kodak.com!psinntp!psinntp!dg-rtp!sheol!throopw
- From: throopw@sheol.UUCP (Wayne Throop)
- Newsgroups: talk.origins
- Subject: Re: Lionel, Evolution and the Theory of Games
- Summary: a additional example
- Message-ID: <721878645@sheol.UUCP>
- Date: 16 Nov 92 00:15:22 GMT
- References: <BxMG96.FHK@usenet.ucs.indiana.edu>
- Lines: 34
-
- : From: lking@athena.mit.edu (Loren King)
- : Message-ID: <1992Nov11.220916.4256@athena.mit.edu>
- : Well, the subject line sort of says it all. Robert Axelrod, William
- : Hamilton and John Maynard Smith have done what you've suggested,
- : although with a different game (i.e. the prisoner's dilemma).
- :: From: adpeters@sunflower.bio.indiana.edu (Andy Peters)
- :: Message-ID: <BxMG96.FHK@usenet.ucs.indiana.edu>
- :: Actually, for even more realism, insert a probability function (as
- :: opposed to a deterministic one): if you lose 50% of the time, you
- :: have a 50% probability of death after each game, etc.
-
- I think an even more realistic scenario exists, which is also closer to
- Lionel's requirement that actual running code be vulnerable to mutation.
- That's the "Core Wars" game, where automatons compete for memory space,
- and try to "eat" their competitors' code. This is more exactly what
- Lionel was asking for: a case where the code itself is blasted, and yet
- improved versions occur. Further, the code for Core Wars is redundant,
- hairy, and armored in various arcane ways, just as living DNA is. As I
- say, a more realistic scenario, and one that bears out the plausibility
- of evolution, even if it is deterministic (sort of) and its analog of
- "sex" is more akin to what "It" does to the men and dogs in the
- antarctic campground, and it may beg the question of abiogenesis,
- but there it is.
-
- Now, I don't know much in detail about it, never having entered Core
- Wars play, but I'm pretty sure of the essentials: 1) a population of
- competing automatons, 2) the automatons are subject to replication
- error, 3) the end result is evolution of this population, and even the
- creation of some better adapted automatons as a result of (1) and (2).
-
- All this, and it wasn't even intended as an example of evolution
- in action (as far as I know).
- --
- Wayne Throop ...!mcnc!dg-rtp!sheol!throopw
-