Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.lang.python > #39887

Re: FYI: AI-programmer

From Robin Becker <robin@reportlab.com>
Subject Re: FYI: AI-programmer
Date 2013-02-25 12:35 +0000
References <4CACF5BFEFC24638A0D8007921722FB8@dev.null> <CAPTjJmq4NNoR9wcq9KSjWOd_bu8gdQ8iuwNVGK-qy1niOJUh_A@mail.gmail.com> <CALwzidnmi+08qGOdSK+=TE5N_38uJ91g468piQ3L2eu5Dzf4kw@mail.gmail.com>
Newsgroups comp.lang.python
Message-ID <mailman.2500.1361795733.2939.python-list@python.org> (permalink)

Show all headers | View raw


On 22/02/2013 19:21, Ian Kelly wrote:
.........
>
> Indeed, it seems to me that this is basically Richard Dawkins' weasel
> program, with the addition of a transformation step in the fitness
> function that amounts to running the string through a Brainfuck
> interpreter.  There is a rather large gap between this and getting
> computers to generate programs that do anything interesting.
......
this has been done before over and over

eg

http://www.ooblick.com/software/evolve/
http://hampshire.edu/lspector/push.html

I think if the OP could get his brainf*ck algorithm to produce a self printing 
program in brainf*ck that would be quite amusing, but for real advances I 
suggest he makes it generate itself. His fitness is fairly silly since it has no 
idea of fitness except success; minimizing the cpu/memory/program length etc etc 
are all additional measures that could be added/used.
-- 
Robin Becker

Back to comp.lang.python | Previous | Next | Find similar | Unroll thread


Thread

Re: FYI: AI-programmer Robin Becker <robin@reportlab.com> - 2013-02-25 12:35 +0000

csiph-web