The battle over cumulative selection and Dawkins’ Weasel program has raged on for some months [years?] here at TSZ and across numerous threads. So can it possibly be that we now, finally, have a definitive statement about cumulative selection?
Mung: And whether or not my program demonstrates the power of cumulative selection has not been settled…
To which keiths responded:
keiths: Anyone who understands cumulative selection can see that it doesn’t, because your fitness functions don’t reward proximity to the target — only an exact match. The fitness landscapes are flat except for a spike at the site of the target.
So there you have it. You need a target and a fitness function that rewards proximity to the target.
Imagine my surprise when I discovered that I had said the exact same thing nine months ago.
Mung: Here’s what he Weasel program teaches us:
1.) In order to demonstrate the power of cumulative selection one must first define a target.
2.) In order to demonstrate the power of cumulative selection one must define a fitness function that increases the likelihood of the search algorithm to find the target relative to the likelihood of a blind search finding the target.
Now perhaps I have misunderstood keiths here. Perhaps he did not really say, or really mean, what I think he said, or what it appears like he said. So I’d like to hear his response.
Is it possible that keiths has agreed with me all along while expending every effort possible to make it seem otherwise?
Just so there’s no mistake, here he is again saying the same thing:
keiths: Mung,
Besides failing in your attempt to code a Weasel and contradicting yourself regarding your intent, you also failed to demonstrate the power of cumulative selection in your program.
1) Your program doesn’t evolve a phrase; it evolves individual letters, one after the other, latching each one when it matches.
2) There is a separate fitness function for each letter.
3) The fitness functions don’t reward proximity to the target — they only reward an exact match for a single character.
The only thing your program demonstrates the “power” of is latching, not cumulative selection.
It’s a remarkable display of incompetence.
Perhaps. But it served its’ purpose. keiths admits I was right all along. So incompetence? Perhaps not.
You need a target. You need a fitness function that rewards proximity to the target. Is that your story keiths, and are you sticking to it? Weasel out of this!
I predict keiths will try to make this about my program and what it does or does not demonstrate rather than his revelation about cumulative selection.
All you have to do is read the comments keiths has made about me to understand that one. Or, I’m not being contrarian, I am merely being skeptical. 🙂
Atheists, above all, are supposed to be reasonable, giving reasons for what they believe, unlike us pathetic creatures of faith. I am trying to understand the belief that some folks have in “the power of cumulative selection.”
I’m willing to discuss that in a friendly manner with anyone.
If you think the biomorph program demonstrates the power of cumulative selection I am going to ask why you think that. I don’t think it does. I think that Dawkins asks us to make a leap from Weasel to biomorphs that he has not actually justified.
Yes, we can select a shape we like and use that shape as the basis for the next round of mutation/selection. And yes, something like that may be going on in biological evolution. But we’ve lost track of the probability argument where Dawkins started. Or are they no longer important?
The power of humans to shape events is not in question. What happens when you remove the human element? What then provides the power of cumulative selection?
Says who? I don’t agree with that statement, it is a terrible caricature of relgious people and I know plenty of stupid atheists. Really, really stupid atheists.
What is that belief, can you quote them stating that belief?
If they are mistaken in that belief, what does that entail?
Does it mean that cumulative selection does not take place?
Or that cumulative selection does not contribute to, or has not contributed to the diversification of life?
I already explained that. Pick some arbitrary trait and select for it, it is usually incrementally enhanced when you do that.
The biomorphs program is not the best possible demonstration of cumulative selection to be sure, yet it does show what happens when you cumulatively apply selection for the same train: It is enhanced.
Why?
A leap in what sense?
I’m not sure what probability argument you refer to, though I admit I never was much interested in this whole thing about computer programs so I might have just missed it.
Are you referring to the difference between just doing completely random scrambles of letters, versus doing something like selection?
The environment does. The environment is a very broad term that really encompasses everything that affects reproductive success of the organism.
It’s just that in these types of super simple programs there isn’t any literal environment, so we have to come up with something analogous to an environment. An abstraction of it, because programming a simulated environment is laborious, complicated and also very computationally demanding to actually run.
By doing something like a target sequence and programming a simple fitness function all of those practical difficulties are dispensed with. But why would you think there is anything lost there that somehow invalidates the analogy to a real natural selection pressure?
However a real or simulated environment turns out to be, there’s going to be matters of fact about what combinations of traits are better suited for survival and reproduction IN that environment, than others. So it just follows logically that, in so far as reproducing organisms exist in that environment, and they are subject to mutations when they reproduce, the environment will either enhance or repress some traits over time. Do you really doubt the truth of this?
Wesley R. Elsberry:
What is a complete set of alphabetic characters and who decides what is “proper” in setting up the problem space?
Dawkins seems to have limited his problem space to uppercase ASCII ‘A’ through ‘Z’ and the space character. Surely this is a limited subset of alphabetic characters and was rather arbitrarily chosen at that!
So my choice was more limited than Dawkins. So what. Would Wesley be complaining if it was more expansive?
What is the “proper” size of the search space in order to demonstrate the power of cumulative selection?
The bit about the character set above reminds me of another amusing Mung screwup, paraphrased here:
Paraphrase is always best I say!
ME:
Demonstrating, rather obviously, that I know the difference.
Good grief keiths. You’d make even Jesus blush.
Mung,
Who said anything about whether you know the difference?
Jesus is embarrassed when atheists tell the truth?
Yes.
Mung,
Not all creatures of faith are pathetic.
What is the “proper” size of the search space in order to demonstrate the power of cumulative selection?
Anyone?
It depends on how much power you want to see demonstrated.
Are you assuming a single target? Can’t you scale the number of targets ti the size of the search space and get the same result?