Ahh so you have a special category for software that meet your demands, such that no software can meet your demands, because then it is by definition outside of your category.
Nothing reproduces then. Everything that reproduces takes something else and makes it into something very much like itself.
You seem not to understand what reproduction is now.
2 Likes
Joe tries so hard to hide his sock puppets. Then his angry science-free stock phrases always come out and give him away. 
1 Like
Avida is not a genetic algorithm- that is a fact. So I donât understand your comment.
" Avida is an artificial life software platform to study the evolutionary biology of self-replicating and evolving computer programs (digital organisms)."
Just say anything, eh? You seem not to understand what an argument is.
You can call it what you want. It does what you want of it, it implements evolution by natural selection without any goals.
Whether it is technically an algorithm seems to be besides the point.
That quote does not substantiate your claim that Avida doesnât qualify for being an algorithm.
1 Like
Here is the description of Avida from the developerâs web site
An auto-adaptive genetic system, Avida serves as a platform for studies in digital evolution and artificial life. Avida runs populations of self-replicating digital organisms in a virtual world, exposing them to configurable mutation and environmental effects. Through selective pressures, these organisms change and evolve.
Sure sounds like an evolution simulator to me.
3 Likes
No that would actually be you, who failed to recognize an reductio ad absurdum. Not only is it an argument, thereâs a name for the specific type.
Ahh, but is it an algorithm? If itâs not an algorithm, then âblind and mindlessâ evolution is somehow false and canât be simulated on computers, or whatever nonsense our renamed retired female marine biologist (aka toaster and refrigerator repair man) with anger management issues was trying to imply.
1 Like
Quietly waits for JoeG to finish googling for quotes he can triumphantly present to prove that Avida isnât an algorithm.
1 Like
Do you know a guy named Wes Elsberry? Ask him if Avida is a GA. And avida does not mimic natural selection. It does not mimic drift. It does not mimic mainstream evolution.
What type of evolution does it simulate? ID only argues against materialistic evolution- That of Darwin, Mayr, Dawkins and Coyne- ie mainstream evolution. The following is from Lenski:
Blockquote âAt the other extreme, 50 populations evolved in an environment where only EQU was rewarded, and no simpler function yielded energy. We expected that EQU would evolve much less often because selection would not preserve the simpler functions that provide foundations to build more complex features. Indeed, none of these populations evolved EQU , a highly significant difference from the fraction that did so in the reward-all environment (P ~= 4.3 x 10-9, Fisherâs exact test).â
What is Intelligent Design and what is it challenging
HINT- It ainât mere evolution.
Thatâs not a fact.
Evolutionary algorithms - or rather selection functions - vary from comparisons to specific goals (e.g. weasel), via comparisons using specific criteria (e.g. Steiner tree length, boxcar distance, soft cube model speed, time-seeking ability), comparisons using less specific (e.g. winning at draughts or backgammon) or changing criteria (e.g. my path program) and subjective comparisons (e.g. biomorphs) or even no criteria (e.g. face drift), to pure reproductive ability (e.g. Tierra).
Itâs trivial to write an evolutionary algorithm that has no goal whatsoever but simply models genetic drift away from an initial state.*
Thatâs not a fact either.
One genetic algorithm used to generate exam timetables and room allocations often failed to achieve itâs goal of having no cases where an exam had to have multiple sittings simply because it was not possible to do so.
The Steiner tree GA Iâve referenced didnât and couldnât produce the optimum solution for most input configurations simply because additional node positions were limited to integral co-ordinates.
And in one well-known example, a GA intended to produce a tone-generating circuit produced an aerial receiver and amplifier instead, which failed completely when run in its intended environment.
Thatâs not a fact either. Many, possibly most, GAâs only include some means of comparing two or more âorganismsâ and picking one or more to âreproduceâ. Avida and Tierra donât even do that much.
Thatâs a completely different topic, but itâs no more reliable than any of your previous âfactsâ.
Which brings me back to the question you didnât answer: Have you ever written or experimented with a genetic algorithm?
Since you seem completely unaware of the range of GAs that have been produced, or the different options that might be available when implementing or running them, the answer appears to be a clear ânoâ.
*Just tweaked my path generator to have no goal and to determine randomly whether an âorganismâ or one of its competitors goes through to subsequent generations. It still runs, and still produces ever-more complex âorganismsâ, even though they ignore the goal completely. Just compare these two pictures:
Iâve just demolished your claim that evolutionary algorithms are goal-oriented by writing one that isnât.
1 Like
It is the claim of computer science that GAs are goal oriented. And yes I have written and experimented with GAs. And the one you just wrote doesnât produce FI. So you lose.
Avida and Tierra donât even do that much.
Neither of those is a GA.
Yes.
Are you unable to find such work? Do you need help constructing search queries?
1 Like
@moderators, we are far affield. I suggest closing and splitting.
The whole world is unable to find such work. If such work existed people who did it would have won a Nobel Prize for their work.
So please, go ahead and post the references. The world awaits and has been waiting for over a century.
I have no problems doing so. On what basis do you speak for the whole world?
No one can find such work, John Mercer. That includes the whole world.