|
Register | Sign In |
|
QuickSearch
EvC Forum active members: 64 (9164 total) |
| |
ChatGPT | |
Total: 916,902 Year: 4,159/9,624 Month: 1,030/974 Week: 357/286 Day: 13/65 Hour: 0/1 |
Thread ▼ Details |
Member (Idle past 3403 days) Posts: 301 From: Burlington, Canada Joined: |
|
Thread Info
|
|
|
Author | Topic: Evolution as an Algorithm | |||||||||||||||||||||||||||||||
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.
|
|||||||||||||||||||||||||||||||
nwr Member Posts: 6412 From: Geneva, Illinois Joined: Member Rating: 4.5 |
I would like to ask you to expand on a couple of your comments.
The technical definition of an algorithm is in terms of a Turing machine. You place the data on the TM tape. The TM operates, and the answer is what remains on the tape. If you are not familiar with the Turing machine, we can describe it in terms of a computer. You enter data at the keyboard. Then the algorithm operates on that data. Finally the result is displayed. One significant point here is that an algorithm, by definition, is non-interactive. You may interact with the computer by providing data, but that's before the algorithm begins to operate. You may interact again, reading the answer, but that is after the algorithm has completed. There is no interaction during the algorithm. Now consider the windows computer. You move your mouse just a little. The computer applies an algorithm and changes what appears on the screen. That's algorithmic, again with no interaction from the time the algorithm starts to the time that it completes (by updating the screen). So the windows computer uses algorithms. But if we look at it overall, then the computer is reacting to your mouse movements and updating the screen. But you watching the screen and your mouse movements are your reaction to what you see changing on the screen. This overall activity is highly interactive, with the computer reacting to you and you reacting to the computer. Since it is interactive, it is not algorithmic (even though it uses algorithms). Biological systems are far more interactive than that. Evolution involves all sorts of interactions. So it, too, is not algorithmic. For me, this mutual interaction is an essential part of evolution (and of a computer operating system). So it is not just a quibble that I claim it is not an algorithm.
Also, I don't see how evolution being an algorithm validates the "prescribed" hypothesis.
An algorithm is deterministic, so it leads directly to the kind of determinism that JAD assumes in his hypothesis. Interaction is not obviously deterministic, although it could admittedly be (as some argue) non-obviously deterministic. My point is that if you assume evolution is an algorithm, then you are pretty much assuming some sort of determinism. Incidently, at least part of the arguement between Dawkins and Gould has to do with determinism. Dawkins is a biological determinist, and Gould takes the opposing view. All biologists will admit that both biology and environment play some role in behavior. But a biological determinist such as Dawkins tends to emphasize the biology and genetics, while Gould looks for other possible influence. For example, if there is a discussion about altruism, Dawkins will tend to seek an explanation in genetics, while Gould would have looked for an explanation in the choices of the individuals that have altruistic behavior.
I rather liked the book, although ...
Dennett is a pretty good writer. I do enjoy reading his work, even when I disagree with it. But, in the case of DDI, I was a bit turned off by his emphasis on biological determinism and his attack on Gould.
|
|||||||||||||||||||||||||||||||
nwr Member Posts: 6412 From: Geneva, Illinois Joined: Member Rating: 4.5 |
Is it perhaps cybernetic then given the importance of mutual interaction, i.e. feedback (Is there a gaian in the house?).
"Cybernetic" fits much better than "algorithmic".
Aren't there algorithms which accept inputs from random sources?
Sure. But the input is not the algorithm. The term "algorithm" applies to the sequence of well defined steps applied to the input.
Why could an evolutionary algorithm not be a probabilistic Turing machine?
Once you add a random number generator, it is no longer a Turing machine. It would be better to say "evolutionary process" instead of "evolutionary algorithm".
|
|||||||||||||||||||||||||||||||
nwr Member Posts: 6412 From: Geneva, Illinois Joined: Member Rating: 4.5 |
I wonder if we could look at evolution as a sort of nested algorithm? That is, repeated cycles of reproduction-with-variation-plus-selection, with the output of one cycle feeding the next and change in the selection rule from time to time.
Yes, that woud be a better description. There is another point that I didn't mention in my earlier posts. An algorithm makes its decisions based on a predetermined standard. When we use an algorithm in a computer, we describe it as using logic, and we describe the decisions as being made on the basis of truth or falsity. Evolution, on the other hand, works with pragmatic judgement - does an organism survive. To redescribe that, an algorithm is a logic process that makes decisions by applying a predetermined standard. Evolution is more of a measuring process that samples reality to see what works as its way of making decisions.
|
|||||||||||||||||||||||||||||||
nwr Member Posts: 6412 From: Geneva, Illinois Joined: Member Rating: 4.5 |
In what sense is Parasomnium's genetic algorithm deterministic?
Does the term "genetic algorithm" apply to the individual steps taken, which are algorithmic and deterministic? Or does it apply to the overall process which uses random input, and is thus neither algorithmic nor deterministic (unless a deterministic pseudo-random number generator is used)?
|
|||||||||||||||||||||||||||||||
nwr Member Posts: 6412 From: Geneva, Illinois Joined: Member Rating: 4.5 |
In the classic definition, you are correct, but that is getting muddy these days, especially with machines receiving input from sensors and even other independent devices.
The "receiving input from sensors" is not part of the algorithm. Rather, the algorithm is in the application of rule based procedures to the input that has been received.
|
|||||||||||||||||||||||||||||||
nwr Member Posts: 6412 From: Geneva, Illinois Joined: Member Rating: 4.5 |
I don't know why you're insisting that an algorithm must be deterministic. Despite the standard definition of an algorithm being something that always generates the same outcome given a fixed set of inputs, in reality an algorithm is just a set of instructions. If one of those instructions says, 'generate me a set of solutions with a certain amount of random input', then what you've got is a non-deterministic algorithm.
I teach that class. The term "non-deterministic algorithm" is a term of art. One must avoid confusing the expression "non-deterministic" as used here, with the philosophic notion of indeterminism. As I am careful to point out to my students, a non-deterministic algorithm is perfectly deterministic. In theoretical use, a non-deterministic algorithm is sometimes described in terms of periodically consulting an oracle for advice on the next move. The advice from the oracle is part of the input to the algorithm. The output of the algorithm is a deterministic function of its input. In practical use, we sometimes replace the oracle with a random number generator, but the algorithm still produces results that are a deterministic function of its inputs. Here the random numbers are part of the inputs. In another form of practical use. we apply a back-tracking or similar method, so that we can systematically try all possibilities. The resulting procedure is completely deterministic.
See Wikipedia entry on Nondeterministic Algorithms
That wiki article even indicates how one would turn the "non-deterministic algorithm" into a deterministic one in order to actually implement it.
|
|||||||||||||||||||||||||||||||
nwr Member Posts: 6412 From: Geneva, Illinois Joined: Member Rating: 4.5 |
As written, the wiki article is a bit ambiguous.
It could mean that the random number generator is to be considered as outside the algorithm, so what the algorithm does is deterministic. Or, as you suggest, it could mean that the random number generator is to be considered inside, but the effect is deterministic in a probablistic sense. I suspect the second meaning (i.e. the one you suggested), is what was intended.
|
|||||||||||||||||||||||||||||||
nwr Member Posts: 6412 From: Geneva, Illinois Joined: Member Rating: 4.5 |
At heart, DNA is made up of a limited number of combinations of base pairs. It would be easy to apply the same logic to the real system since the randomiser that is "mutation" can only work within tightly defined criteria (ie. a finite number of base pairs). For any mutation event during reproduction,
Let's keep in mind that there is more to biology than DNA. People sometimes talk as if the DNA is a specification or blueprint for an organism. But it isn't. Rather, it is specification for manufacturing a variety of proteins. A biological organism comes about when this manufacturing is done in a suitable development environment. We must also remember that our evolutionary history might have involved symbiotic unions. For example, Lynn Margulis has proposed that the eukaryote cell originated as a symbiotic union. If correct, then the structure of that symbiotic union is part of what is passed on to the next generation, and the way that structure is passed on is not as part of the DNA specification.
|
|||||||||||||||||||||||||||||||
nwr Member Posts: 6412 From: Geneva, Illinois Joined: Member Rating: 4.5 |
I'm not sure quite how you work this out.
For example, the mitochondrian is inside the cell, rather than existing as a separate cell. That's part of the structure that is passed on.
|
|||||||||||||||||||||||||||||||
nwr Member Posts: 6412 From: Geneva, Illinois Joined: Member Rating: 4.5 |
From Message 19
You are missing a bit of context there. The "it" in the first sentence you quote from me refers to the claim that evolution is an algorithm. Note that I have no problem with the view that evolution uses algorithmic processes. My objection is to the claim that evolution itself is an algorithm.
An algorithm is deterministic, so it leads directly to the kind of determinism that JAD assumes in his hypothesis. Interaction is not obviously deterministic, although it could admittedly be (as some argue) non-obviously deterministic. My point is that if you assume evolution is an algorithm, then you are pretty much assuming some sort of determinism. The effect of an interaction is deterministic in the sense that you insist on when you're teaching your students about non-deterministic algorithms, i.e. the interaction 'produces results that are a deterministic function of [its] inputs'. This discussion is getting more complicated than it needs to be because you're confusing the philosophical notion of determinism (every event having a physical cause), with the common use of the terms 'deterministic' and 'non-deterministic' to mean 'predictable' and 'non-predictable'.
I can't say that I have ever seen "deterministic" used to mean predictable. Or, if I have, that mistaken notion has quickly been corrected. There are computer programs whose output we cannot predict (except by actually running the program). That does not make the program non-deterministic. "Predictable" has partly to do with our state of knowledge, while "deterministic" does not.
|
|||||||||||||||||||||||||||||||
nwr Member Posts: 6412 From: Geneva, Illinois Joined: Member Rating: 4.5 |
..., and I think that Nwr would agree with me that it is important to separately consider an algorithm from its inputs.
Quite right. I have no disagreement with your use of the terminology. You are correct, that the disagreement between me an JavaMan is mostly semantic. I think my disagreement with Dennett in DDI is more than semantic. It was certainly my impression that Dennett was making too strong a claim.
|
|||||||||||||||||||||||||||||||
nwr Member Posts: 6412 From: Geneva, Illinois Joined: Member Rating: 4.5 |
You might have been reading more into my post than I intended. I don't see anything I disagree with in your latest "refutation."
|
|
|
Do Nothing Button
Copyright 2001-2023 by EvC Forum, All Rights Reserved
Version 4.2
Innovative software from Qwixotic © 2024