|
Register | Sign In |
|
QuickSearch
Thread ▼ Details |
Member (Idle past 3401 days) Posts: 301 From: Burlington, Canada Joined: |
|
Thread Info
|
|
|
Author | Topic: Evolution as an Algorithm | |||||||||||||||||||||||||||||
Woodsy Member (Idle past 3401 days) Posts: 301 From: Burlington, Canada Joined: |
In his book, Darwin's Dangerous Idea, Daniel Dennett outlines the idea that evolution is an algorithm, similar to a computer program. That is, it is a sequence of steps (reproduction with variation then selection) that reliably produces a particular result (evolution). He shows that algorithms are independent of the substrate they run in. For example, some computer programs use the evolution algorithm for design work.
If one can show that the description of the evolution algorithm is correct and if one can show that biological systems do exhibit reproduction with variation, and that selection operates, surely one should then expect evolution to occur. If these requirements are satisfied, is that sufficient for confidence in the ToE? Don
|
|||||||||||||||||||||||||||||
AdminPhat Inactive Member |
Woodsey, welcome to EvC. That sounds like a good book, although I have not read it.
Edited by AdminPhat, : No reason given.
|
|||||||||||||||||||||||||||||
AdminPhat Inactive Member |
Thread moved here from the Proposed New Topics forum.
|
|||||||||||||||||||||||||||||
ramoss Member (Idle past 640 days) Posts: 3228 Joined: |
A computer algorthym might show a model of how evolution might happen, but there is more that has to happen. You need examples in the real world. Evolution has this. You have to have real world data. The algortithm only gives some kind of explaination for the data that is observed. The algorithm in and of itself is not enough. It has to be able to mimic the actual data.
|
|||||||||||||||||||||||||||||
Woodsy Member (Idle past 3401 days) Posts: 301 From: Burlington, Canada Joined: |
Perhaps I was not quite clear. One way of looking at an algorithm is as a "black box" that, if presented with proper inputs, produces a particular output. In the case of evolution, it seems to me the proper inputs would be a) any kind of object that exhibits reproduction with variation and b) some kind of selection rule. The output would be changes in progeny (over time) that accommodate the selection rule, ie evolution.
|
|||||||||||||||||||||||||||||
Hyroglyphx Inactive Member |
I have to agree that more than just an abstract mathematical theorem or a computer simulator is neccesary to to bring the ToE out of its arcane understanding. There are programs that simulate how diversity can arise within a population, none of which anyone is in disagreement about. However, this is precisely why there are multiple threads open on the differences between a macroevolutionary process and a microadaptive process. Furthermore, models tend to be grossly oversimplified to the point that it cannot be cited as a proof in defense of the theory.
As well, computer simulations of an evolutionary process seem to imply that the development which is being measured over so many generations is independent of development of other structures which are necessary for basic function. This, of course, is what IC "Irreducible Complexity" is all about. And perhaps this is what Darwin was alluding to when he said that if it can be demonstrated that nothing could have come together by successive chance, my theroy would utterly break down. Lastly, the changes observed from the simulation are dependent on the original data input which could clearly be construed as biased reserch or leading the research in the direction of the programer desires. When the theory is still lacking evidential credibility, why abandon the field and the lab and grind out abstract theorems or abstruse computer simulations? We need something far more laudable than these to seal the deal on the plausibility of the ToE. “It is in vain, O' man, that you seek within yourselves the cure for all your miseries. All your insight has led you to the knowledge that it is not in yourselves that you will discover the true and the good.” -Blaise Pascal
|
|||||||||||||||||||||||||||||
Woodsy Member (Idle past 3401 days) Posts: 301 From: Burlington, Canada Joined: |
I am not suggesting the use of a computer model here. The argument would run something like: If the algorithm is valid (ie reproduction with variation plus selection yields evolution), then if any system exhibits reproduction with variation, we should expect evolution in the presence of selection. One would still have to verify that the algorithm is correct and that any system of interest does exhibit the required behaviour, reproduction with variability.
Questions such as the scope of the changes etc are another matter, and likely much more interesting.
|
|||||||||||||||||||||||||||||
Parasomnium Member Posts: 2224 Joined: |
Hello Woodsy,
First of all, a welcome to EvC from me too. To anybody else (Phat?) who reads this: I cannot recommend enough the book Woodsy mentioned. It's a wonderful read and in my opinion the best book about evolution by a non-biologist. The idea of evolution as an algorithm is a very powerful tool in thinking about evolution. I have mentioned it often here on EvC that the principle mechanism of evolution has been demonstrated very convincingly in the form of evolutionary algorithms. Being a programmer, I have tried it myself, and seen with my own eyes that the simple repeated application of random variation with selection does indeed result in evolution of whatever is being subjected to this procedure. For me, what you suggest has been demonstrated. Having written an evolutionary algorithm myself, I know that I have accurately implemented random variation with selection. I have seen evolution to be the result. I am totally convinced. "Ignorance more frequently begets confidence than does knowledge: it is those who know little, not those who know much, who so positively assert that this or that problem will never be solved by science." - Charles Darwin. Did you know that most of the time your computer is doing nothing? What if you could make it do something really useful? Like helping scientists understand diseases? Your computer could even be instrumental in finding a cure for HIV/AIDS. Wouldn't that be something? If you agree, then join World Community Grid now and download a simple, free tool that lets you and your computer do your share in helping humanity. After all, you are part of it, so why not take part in it?
|
|||||||||||||||||||||||||||||
ramoss Member (Idle past 640 days) Posts: 3228 Joined: |
You are also drawing upon the vast amounts of evidence from the real world for your conviction though.
The algorithm in and of itself shouldn't convince you. It is the algorithm that fits into what we observe in the real world that makes it a strong statement.
|
|||||||||||||||||||||||||||||
Hyroglyphx Inactive Member |
Being a programmer, I have tried it myself, and seen with my own eyes that the simple repeated application of random variation with selection does indeed result in evolution of whatever is being subjected to this procedure. For me, what you suggest has been demonstrated. Having written an evolutionary algorithm myself, I know that I have accurately implemented random variation with selection. I have seen evolution to be the result. I am totally convinced. Did you take into consideration my objections? Aside from my previous objections, I think you would be able to appreciate that no amount of algorithmic or logarithmic expressions could ever substitute actual empirical data. Perhaps you are looking at the substrate, the material, and the very clinical definitions of what these algorithms are able to produce. The way it makes it happen, there are no catastrophic variables that would throw these readings in a tizzy. But as we should all know quite well, nothing in life is free, and things don't organize themselves by chance innumerable times. “It is in vain, O' man, that you seek within yourselves the cure for all your miseries. All your insight has led you to the knowledge that it is not in yourselves that you will discover the true and the good.” -Blaise Pascal
|
|||||||||||||||||||||||||||||
Parasomnium Member Posts: 2224 Joined: |
nemesis_juggernaut writes: models tend to be grossly oversimplified to the point that it cannot be cited as a proof in defense of the theory. This is not at all true. The beauty of the mechanism of evolution is that it can be implemented quite accurately on a computer. There is no need to simplify, let alone oversimplify, the principle of random variation. If you have data in the memory of a computer and you write a subroutine that makes copies of those data while randomly changing parts of it, then you have implemented reproduction with random variation very exactly. Next, if you make a function that determines the fitness of the data copies and filter out the less fit for the next round of reproduction, you have implemented selection very exactly. That's all there is to it: implement reproduction with random variation, implement selection, and let it run for a while, i.e. let the procedure repeat itself for a significant number of generations, and you'll see the data evolve. I must also stipulate that a built-in goal or direction of any kind does not figure anywhere in the programme. It is not necessary, evolution will take place regardless.
As well, computer simulations of an evolutionary process seem to imply that the development which is being measured over so many generations is independent of development of other structures which are necessary for basic function. This, of course, is what IC "Irreducible Complexity" is all about. Evolutionary algorithms do not just imply this independence, they actually prove it. And thus they also falsify 'irreducible complexity'.
the changes observed from the simulation are dependent on the original data input which could clearly be construed as biased reserch or leading the research in the direction of the programer desires. It could perhaps be so construed, but that doesn't mean it's true. The only way to find out is to do it yourself. Which is what I have done, and which is what you should do. I guarantee you that the experiment I have outlined above is repeatable. If you program it yourself you can be absolutely sure that your programme works the way you intended. (If you're a good programmer, that is.) And if you have faithfully implemented reproduction with random variation and selection, and nothing more than that, then I guarantee you that you'll see evolution happening before your very own eyes.
When the theory is still lacking evidential credibility, why abandon the field and the lab and grind out abstract theorems or abstruse computer simulations? OK, don't trust other programmers. There can be nothing abstract about it, or abstruse, if you try it yourself. "Ignorance more frequently begets confidence than does knowledge: it is those who know little, not those who know much, who so positively assert that this or that problem will never be solved by science." - Charles Darwin. Did you know that most of the time your computer is doing nothing? What if you could make it do something really useful? Like helping scientists understand diseases? Your computer could even be instrumental in finding a cure for HIV/AIDS. Wouldn't that be something? If you agree, then join World Community Grid now and download a simple, free tool that lets you and your computer do your share in helping humanity. After all, you are part of it, so why not take part in it?
|
|||||||||||||||||||||||||||||
Parasomnium Member Posts: 2224 Joined: |
ramoss writes: You are also drawing upon the vast amounts of evidence from the real world for your conviction though. Of course I am. I never suggested that I was convinced of the truth of evolution by the evidence of evolutionary algorithms alone. But they are a very convincing argument about the principle mechanism of evolution though. "Ignorance more frequently begets confidence than does knowledge: it is those who know little, not those who know much, who so positively assert that this or that problem will never be solved by science." - Charles Darwin. Did you know that most of the time your computer is doing nothing? What if you could make it do something really useful? Like helping scientists understand diseases? Your computer could even be instrumental in finding a cure for HIV/AIDS. Wouldn't that be something? If you agree, then join World Community Grid now and download a simple, free tool that lets you and your computer do your share in helping humanity. After all, you are part of it, so why not take part in it?
|
|||||||||||||||||||||||||||||
Parasomnium Member Posts: 2224 Joined: |
The way it makes it happen, there are no catastrophic variables that would throw these readings in a tizzy. But as we should all know quite well, nothing in life is free, and things don't organize themselves by chance innumerable times. I'm sorry, I don't quite get what you mean by that first sentence. Could you please explain?
things don't organize themselves by chance innumerable times You are making an oft repeated mistake: you leave out selection. Evolution is not just a jumble a random events. It's the non-random selection that produces the interesting results. "Ignorance more frequently begets confidence than does knowledge: it is those who know little, not those who know much, who so positively assert that this or that problem will never be solved by science." - Charles Darwin. Did you know that most of the time your computer is doing nothing? What if you could make it do something really useful? Like helping scientists understand diseases? Your computer could even be instrumental in finding a cure for HIV/AIDS. Wouldn't that be something? If you agree, then join World Community Grid now and download a simple, free tool that lets you and your computer do your share in helping humanity. After all, you are part of it, so why not take part in it?
|
|||||||||||||||||||||||||||||
nwr Member Posts: 6412 From: Geneva, Illinois Joined: Member Rating: 4.5 |
It looks as if there is going to be disagreement on this thread.
No, evolution is not an algorithm. To give a little perspective, I'll add that the Windows operating system is also not an algorithm (though it makes use of algorithms). There is a technical mathematical definition of "algorithm", and neither evolution, nor the Windows operating system, meet that definition. I didn't much like Dennett's book either. In my opinion, S.J. Gould's understanding of evolution is better than that of Dawkins. But Dennett sided with Dawkins and attacked Gould in DDI. I guess I could put it this way. If evolution is an algorithm, then JAD's Prescribed Evolutionary Hypothesis is correct.
|
|||||||||||||||||||||||||||||
Hyroglyphx Inactive Member |
quote: I'm sorry, I don't quite get what you mean by that first sentence. Could you please explain? My apologies. In retrospect I barely understood it myself. Allow me to clarify. What I meant to say is that I doubt variables, such as catastrophes are not introduced as possible scenarios that would effect the outcome. In other words, it makes the presumption that life just kinda organizes itself beneficially without assistance. If ever there was one single argument that is truly presenting a problem for pro-evolutionists is IC. Are these evolutionary programs advanced enough to factor in such improbablities? I say that I doubt it because I've seen a few evolution simulators that are about as complex as "Pong."
You are making an oft repeated mistake: you leave out selection. Evolution is not just a jumble a random events. It's the non-random selection that produces the interesting results. How did you determine that I leave out selection? There is only so far the genes can drift in a population before you hit a brick wall. The fallacy is that the genome of any given organism is basically infinite in its variability. That's obviously not the case. Edited by nemesis_juggernaut, : typo “"All science, even the divine science, is a sublime detective story. Only it is not set to detect why a man is dead; but the darker secret of why he is alive." ”G. K. Chesterton
|
|
|
Do Nothing Button
Copyright 2001-2023 by EvC Forum, All Rights Reserved
Version 4.2
Innovative software from Qwixotic © 2024