I was very struck by Glenn Williamson’s [vjt meant GlenDavidson] remark that creativity is not the same thing as complexity. Very deep. Glenn seems to think that people are good at the former, but the blind processes can outdo them in the latter. That’s an interesting view, but I’d want to see evidence that blind processes are actually capable of producing systems with a high degree of functional complexity, of the kind Axe described in his book. Even a computer simulation would be something.
What with all the experts in writing GA’s here at TSZ I was hoping VJT would have elicited more of a response.
Which brings me to my second point. Tom English remarked above:
“What’s more important in responding to Axe, I suspect, is the issue of knowledge. Do you get only what you know how to make? The answer to that, coming from evolutionary computation, is a top-of-the-lungs NO!!!. I’ve set up a number of evolutionary systems that ended up knowing, in clear operational terms, how to do what I hadn’t a clue how to do.”
I’d like to ask Tom: in terms of building functional coherence, what’s the best your algorithms are capable of doing? I’m interested in finding out, and if you can point me to a good place to start familiarizing myself with GAs, I’d be grateful.
Why doesn’t someone walk VJT through designing a GA that does what he is asking? Is it because a generic GA doesn’t do anything of the sort of thing he is asking for?
For VJT, do you have any experience at all in any programming language?
You’re accusing me of lying now, so bugger off. Go drink the #KeithsKoolaid. What is it with some of the members here, lol. God I miss hotshoe_.
Of course you don’t. Because You know full well that I am willing to put up 10,000.00 dollars in a wager that I can write a GA and that you and no one else here will be willing to match that wager.
Given the level of certainty some of you seem to have about my abilities you should even be offering me odds.
But you won’t. And that’s another reason for me not to waste my time on you and others like you who have nothing better to do than accuse me of being unable to do so.
Look, just get ten of you “skeptics” to put up 1 grand each. We’ll set up an escrow account. Joe F and I can meet up down at the DI offices in Seattle if you trust him with your money. Start a GoFundMe. I’ll even give your 10k to the Indian children in Oklahoma in memory of “The We Should have Known Better” club after I’ve won the wager. It’s a win win.
I’m willing to put my money where my mouth is, as the saying goes. And you’re not. So here’s another saying for you: Money talks, and bullshit runs a marathon.
Mung,
So predictable and tragic.
Wah wah you’ve accused me of lying, I want a bet. Barry’s boring flying monkey.
Surprisingly, boring seems to be a worse sin than flaming.
If we can’t have NewMung, could we at least have InterestingMung?
Call me MungWithMoney. If that fails to interest you, I think we all know why.
You and Richardthughes ought to be able to come up with ten grand between the two of you, if you really think it’s worthwhile.
But why aren’t you offering odds, given your belief that I am just bluffing?
I have a theory about that.
So many skeptics, so little cash. LoL.
This is a serious offer. I am willing to put up ten thousand dollars as a wager against those who claim I am incapable of coding a GA.
These “skeptics” should be offering me odds, but maybe I need to offer them odds. What odds do you “skeptics” need in order to feel that your wager has a chance of netting you any money?
Think of the children.
Oh, perhaps the stakes are not high enough. What stakes do you “skeptics” want to play for?
I think it’s you “skeptics” who are bluffing, and that you believe it’s better to bluff when there is nothing at stake. But then, what’s the point of that?
Anyone here actually willing to put real money on the line? I promise, your losses will go to a good cause.
You will code a GA that solves a problem of our choosing in a controlled environment?
If you define what you think qualifies as a GA and can substantiate your claim that a GA must solve a problem of your choosing in order to qualify as a GA.
I claim that I can code a GA. I don’t claim that I can code a GA that solves any problem. In fact, I have rather consistently called into question that there is any such GA.
Submit your problem to be solved and I will tell you whether I think I can code a GA to find a solution and whether I am willing to wager on whether I can write a GA to solve that specific problem.
What is your confidence level in a claim that I cannot code a GA? Is it hovering near zero? What odds do you want, 10,000 to 1?
Mung,
Lulz. There are kids that recite from the Koran but can’t actually read it. You’ll be supplied on the day with a problem that has already been solved by GAs. Having ‘written by Mung’ in the code won’t cut it. You will code in supervised conditions.
Richardthughes, do you honestly believe that I cannot produce the code for a genetic algorithm? Surely such code can easily be found on the internet.
Perhaps you think that you can easily identify whether the code I produce is my own code or the code of another? How so?
How will you know that I wrote the code unless we set up conditions which you think will be a reliable indicator of whether or not I wrote the code?
These are reasons why I don’t simply post the code. How would you know that I was the one to write it?
Agreed. So please drop the pretense that simply posting code on GitHub would settle the matter. My failure to post code on GitHub means what, exactly?
If you were to post code now we’d have it and could look at it. No problem. You don’t appear to. Coding does not mean ‘have my friend do it’ or ‘copy some trivial example I don’t understand’. I’m just basing my belief in your inability on your previous dipshittery on the subject. Don’t get all pissy. I’m laughing at your fighting retreat, though.
/edited.
I’m actually offering you odds when you ought to be offering me odds.
Retreat. Sure. Whatever.
I’ve never personally met Joe Felsenstein, but I think he might qualify as a referee. I have met Allan Force, but I don’t know if he is still in the Seattle area.
Is there any question about whether these two are qualified to judge whether a computer program qualifies as a GA?
Mung,
The solution to the problem would count.
So? LoL.
If I give you code in one programming language, how do you determine that I didn’t just take code that someone else wrote in some other programming language and convert it into a different programming language?
You think you could spot that? Go ahead, give me a reason to believe that.
Mung,
You could always tell us your inspiration, what you were researching, why you made certain choices. Imagine quizzing an author and an impostor on a book. Do you think their understanding would be the same?
And don’t you think that discovering IP theft related to software already has some tools available?
Post the odds you’re offering, Richardthughes.
Do you doubt that I am willing to wager ten thousand dollars that i can write a GA?
do you agree to the conditions?
keiths and Richardthoughes can fly to Seattle to observe in person and we’ll subtract their plane fare from their wager. It’s still for a worthy cause.
I will wager 10k of my own money, and if I win whatever amount Richardthughes and keiths wager will go to the children. Better for the children if they offer odds.
I can code a genetic algorithm.
No, the terms .
The claim is that I cannot code a genetic algorithm. What other conditions obtain?
The claim is that I cannot code a genetic algorithm. The terms are the conditions under which you can satisfy your claim. What are your terms?
Don’t worry Mung. I believe you could write a GA if you wanted to. I have faith.
Mung,
Depends what you mean by halting. There is nothing about the basic operation of a GA algorithm that makes it halt. If a programmer determines a stop condition, and it hits it, it halts. Or, in the case of the M&M version sans mutation, when people get bored of seeing the same colour never changing and they close the window, it halts. Even when the display stops changing the program is still churning away, you just don’t see anything.
But a GA in operation is not determined by whether it stops at some point in the future, but simply whether it applies operations to populations of strings.
You’d have to word-game ‘selection’ for that to be true. Selection in biology means something other than simple random birth and death, uncorrelated with genetic attributes.
They are ‘picked’ completely randomly. In essence, the selection coefficient of all alleles is zero. There is a case for saying that natural selection includes the neutral case, but more often in biology it is relegated to the non-neutral, non-drift component, for historic reasons.
Mung,
Hey, I’ve got that book! Read it, too.
https://www.sciencedaily.com/releases/2016/09/160902142228.htm
The Designer works in mysterious ways.
You know, when you start to channel Kent Hovind, it might be time to find a new barrel in which you can sink to the bottom.
Where I grew up, a handshake was sufficient. But that was a time when honor was a word associated with something besides killing women.
But that’s beside the point. Our wagerer has not been heard of since he was asked to accept reasonable terms for judging his abilities.
Some people just won’t trust Jesus to keep him honest.
Infidels!
Glen Davidson
Perhaps he thought he could win $10,000 by regurgitating a weasel. Some toy thing whose code could be memorized rather than understood.
My name has been invoked as a possible judge of this wager.
While I’d be happy to meet you folks, and my lab is available as a possible venue, I am hesitant to be a judge for two reasons:
1. Suppose that the matter were close — Mung almost, but not quite writes a GA. It is then up to me to judge, and everyone would be quibbling. Anything I decide would be very unpopular. No thanks, let someone else reap the hostility.
2. Is this wager about anything important? It’s not about what GAs can or can’t do, or whether they do or don’t model evolution, or even a semantic issue like whether they are or are not “searches” or are or are not “directed” or “intelligent”. It’s about whether Mung can or cannot write a GA. Which is important … why? To validate or invalidate opinions Mung has about GAs and evolution?
So, no thanks.
Joe Felsenstein,
I agree. Wagering seems to be the creationist go-to chest thump. I just find it hilarious as one of Mung’s broken records is ‘show us the code’
How about we start with a somewhat less ambitious task?
I would like mung to write out the defining properties of a GA in something like pseudocode.
Shouldn’t take more than half an hour and could be posted right here.
Shouldn’t be copied, though.
The problem I have with all aspects of this wager is that mung continually challenges what it is that a GA does.
Writing out pseudocode clarifies what a program does.
Regarding wagers, I always ask, What Would Jesus Do?
Mung has already demonstrated that he doesn’t understand the weasel algorithm when he posted this bit of code here.
keiths kindly dissected it for those who don’t speak Ruby:
If anyone pursues Mung’s bet, one of the conditions should be that the code be written from scratch on a machine without network connectivity.
Patrick,
Exactly, it’s not that we lack evidence that Mung can code a GA, it’s that there’s ample evidence that he doesn’t know what GA’s are all about.
That could of course have changed since then, and it wouldn’t take him much time or effort to demonstrate he can do it, not much more than it took him to code that anyway.
dazz,
We’re a bit stuck between – “I can write a french paragraph” and “I can speak french”. The first is trivial and doesn’t require you to know or understand French at all.
That’s why I’m not putting money on this one — unlike many UDers (udders?), Mung seems to have the capacity to learn how to code a GA. It’s pretty straightforward to get a simple one running, after all.
If I remember correctly, though, that wasn’t what his original $10,000 wager was about.
Maybe we can discuss someday. It’s certainly been on my radar as a potential topic.
You don’t know how much that means to me. 😉
See, Joe gets it. I’ll post code for a GA when and if I ever feel like doing so. I have nothing to prove here.
Agreed. We just ‘infer to the best explanation’ based on your previous attempts (see above). Numerous threads have been started to help you (and others) understand some of the fundamentals.
Mung:
How did I get dragged into this?
There’s an official weasel algorithm is there?
Did I claim that code is a genetic algorithm? Or was it a demonstration that there’s a faster way to reach the target than what keiths gave us.
And now VJT is asking and I can see just how much people are helping him.
Could you post a link to at least one thread that was started to help me (and others) understand some of the fundamentals? VJT might appreciate it.
A previous exchange: