The Game Theory of Life 85
An anonymous reader writes In what appears to be the first study of its kind, computer scientists report that an algorithm discovered more than 50 years ago in game theory and now widely used in machine learning is mathematically identical to the equations used to describe the distribution of genes within a population of organisms. Researchers may be able to use the algorithm, which is surprisingly simple and powerful, to better understand how natural selection works and how populations maintain their genetic diversity.
Two things (Score:5, Interesting)
1. If the machine learning algorithm has been found to be mathematically identical to the genetic spread algorithm, how would biologists be able to use it to better understand natural selection and genetic diversity? What can they learn from the first algorithm that they couldn't learn from the one they already had? If the two algorithms are mathematically identical, aren't they both just different names for the same mathematical structure? Learning a cat is called neko in Japanese doesn't tell you anything about cats you didn't already know -- it just tells you something about the Japanese language.
2. Are algorithms discovered, or created? If anything is discovered, the underlying mathematical structure more than one algorithm can point to seems to be a better candidate than the algorithms themselves. Fossils are discovered; algorithms are made up.
Re: (Score:2)
Re:Two things (Score:4, Interesting)
succint unproven "fact" for a question that can give work to philosophers for a few years.
are mathematics (of which algorythms are a small part) discovered or created ? No one has a clear answer to that question.
Re:Two things (Score:5, Insightful)
are mathematics (of which algorythms are a small part) discovered or created ? No one has a clear answer to that question.
Really? Maybe it's because the answer is so simple, no one serious has bothered tackling it.
Mathematics is a language. As such, it is created.
The things that mathematics describes are where it gets interesting. Much like in other languages, you have tangible things (easily verified as existing independent of the language), intangible things (dreams, emotions, forces) that are generally accepted as existing independent of language. And then you have two classes of things that are not entirely independent.
You have categories or groups. "Animal" is not an intangible thing, because it doesn't describe anything that actually exists, it is a term for a collection of things that exist. The term itself is semantics, but most categories have an objective component that exists independent of language.
The final category is pure language constructs. Rhymes, sentences, grammar, poems, etc. - while you can argue that they are linked to some biological or neurological element of human nature, a rhyme or a poem is very much a language construct and does neither describe a thing nor a group of things, it's a self-referential language construct.
And if you look closely, you find the same in mathematics.
Re: (Score:3)
But saying mathematics is a language and the things described are not part of mathematics, or perhaps they are, and a poem cannot describe something... I feel you have some work to do before being able to convince every
Re: (Score:2)
Begin with https://en.wikipedia.org/wiki/... [wikipedia.org]
From that page:
Although most mathematicians and physicists (and many philosophers) would accept the statement "mathematics is a language", [...bla bla... it's not as simple, but the more we think about it... ] that the distinction between mathematical language and natural language may not be as great as it seems.
There's actually a more specific article [wikipedia.org] right on WP, but as always, never believe anything you read on WP without checking it against other sources.
Re: (Score:2)
Re: (Score:2)
Are laws of nature created or discovered?
Someone has read too much Mage: The Ascension ;-)
Re: (Score:2)
I do not agree that the question is difficult.
I do agree that a lot of people consider it difficult, because they are trapped in category mistakes and cannot properly seperate their levels of abstraction.
Once you get that right, it isn't all that hard anymore. You just need to go beyond the word, into meaning. What is it, exactly, that you mean by the word "algorithm"? Is it
a) the particular formula in particular mathematical notation?
b) the operations described by that formula?
c) the process described by t
Re: (Score:2)
I think it's still hard because you could equally have:
1) Person A who thinks up the algorithm in his head and writes it down hence creating it
2) Person B who independently spots the algorithm executing in a natural process and writes it down, hence discovering it.
An algorithm is a set of instructions, but how you come to getting into your mind and writing down those instructions results in a dividing line between whether it was created or discovered and even that's not simple when there's a debate about wh
Re: (Score:2)
Again, I think you are mixing different levels of abstraction here.
Let's, for simplicity sake, take an extremely simple algorithm as an example: 1+1=2
Person A writes it down as a purely mathematical description, finds out he can draw nice conclusions from it and generally enjoys it in the sphere of pure math. It was created, no doubt.
Person B does not ever spot "1+1=2" anywhere in nature. What person A spots is that if he has one stone in his hand, and he adds a second one, he now has two stones in his hand
Re: (Score:2)
But I don't think anyone talks about creating a description when they're referring to description vs. discovery of an algorithm. Everyone knows that language, which is ultimately what math is, is a creation of the human mind (or is it discovered inside the human mind? I'll let you figure that one out).
What you're referring to as the creation of an algorithm is simply the creation of the description of the algorithm which is a different thing to the creation/discovery of the algorithm. Fundamentally the poin
Re: (Score:2)
(or is it discovered inside the human mind? I'll let you figure that one out).
There's not even an argument there. Current scientific knowledge indicates strongly that our brains are wired for language processes, but not for a specific language.
What you're referring to as the creation of an algorithm is simply the creation of the description of the algorithm which is a different thing to the creation/discovery of the algorithm.
No. You are trying to introduce some kind of strange sideways category. Or maybe it is the word "algorithm" that's causing the confusion here. I've described my view, how about you describe yours?
Re: (Score:3)
You do realize that these two things are not nearly equivalent, yes?
Finding if a combination of words satisfies all semantical and grammatical requirements is not the same as verifying if some combination of symbols has been published before, no matter which language you talk about.
Math certainly is a very special language in that it strictly obeys the rules of logic, and thus can be used to derive and formulate proofs in such a clear and unambigious way that computers can be used for the purpose. I would c
Re: (Score:3)
until someone invents Russian and invents the word ÃþÃ'Ã'ÃÃ').
seriously, slashdot ?
It's 2014, not 1994. Fucking get some Unicode support.
Re: (Score:1)
Re:Unicode (Score:3)
Drifting off topic, but did the infamous Beta in fact get Unicode support?
I mean, look at this tortuous new Beta, did they even bother to put in the Unicode support that people have been screaming for for ten+ years?
Damn we need a mole at Dice. What do they even do at management meetings?
"Let's make a whole new design with 55 changes."
"What about Unicode Support?"
"That's a big word. That's too hard for me. Let's put more videos up instead."
Re: (Score:3)
You mean like Latin didn't have a word for computer or laser or neutron star?
Because words are added to languages when they are needed. Languages are not created in the "a designer sits down and invents it" sense, but in the sense of continuous improvement.
Re:Two things (Score:5, Funny)
Except esperanto. I'm pretty sure someone sat down and invented that one.
Re: (Score:2)
Languages are not created in the "a designer sits down and invents it" sense, but in the sense of continuous improvement.
My first language is FORTRAN and now I speak C++, they are all designed languages, you insensitive clod.
Re: (Score:2)
Re: (Score:2)
Because Mathematics describes _concepts_.
When people's paradigms enlarged to include more concepts they extended mathematics. The exact same thing happened for imaginary numbers, and quaternions.
How does *any* language name and define foreign concepts? I.e pick a language that doesn't have a term for "email" or "computer"
Re: (Score:2)
Re: (Score:1)
Re: (Score:2)
Mathematics is a language. As such, it is created.
The interesting thing about math is that it is a language that reveals underlying isomorophisms, like the one described in TFA. This feature is one of the things that leads to naive people thinking that the math somehow "precedes" the things it describes.
But we see similar isomorphisms in all languages. Consider the "ballad" form of poem. It occurs in incredibly diverse contexts, but the underlying structure is always the same, which means you can sing "Amazing Grace" to the tune of the theme from "Gilligan
Re: (Score:2)
Phenomena manifest in many ways, many perspectives.
The colour "red" is a human phenomena, and the notion of "wavelengths" and "energy" are also something humans see/have/experience as phenomena. You can then use signifiers, signified, words, sounds, etc. to say "red" for what you see in your vision as "red" whilst noting on some instrument the "wavelength".
There is no really real reality beyond your experience of sights, sounds, concepts, etc. -- reality which we can speak of, IF we can't observe it. In thi
Re: (Score:1)
I have a clear answer to that. Mathematics are discovered. If they were invented, they would work exactly as we wanted them to. But they don't. For example, the pythagoreans were convinced that all numbers could be represented as a ratio of integers. But they were proven wrong.
Re: (Score:2)
> are mathematics discovered or created?
Uh, you do realize the question is not mutually exclusive, right?
The answer is: Both
Stop thinking linear, and start thinking multidimensionally.
Re: (Score:3)
"are mathematics (of which algorythms are a small part) discovered or created ? No one has a clear answer to that question."
I thought it was pretty clear that stuff is discovered. To me this kind of question reads like "Well, does it stop being right at any time once it is discovered?" and the answer is generally no. (A, you sometimes get stuff that was discovered and not properly reported, at which point the original discovery is not at fault and it is just a reporting problem, or B, you get stuff that was
Re: (Score:3)
When you have two distinct things, which you understand to different extents, proving that they're identical allows you to learn about one thing from the knowledge of the other thing.
To use your example:
Prior to today, we knew that cats lapped up milk with their tongues, and also preen their fur with their tongues. Also prior to today, we knew that a Japanese animal called neko coughs up balls of stuff.
Today we found out that cats are identical to neko.
We now know that cats cough up balls of stuff, and that
Re: (Score:1)
Just call it "Intelligent Design". Lots of people are going to be claiming they were right all along, although most of them won't have any chance of actually understanding the maths.
Equations, not algorithm (Score:2)
Re: (Score:2)
Because names are properties acquired by stipulation. Discovery entails an existence of algorithms independent of humanity, while creation entails the existence of algorithms depends on humanity (or other algorithm creators). That's why it's a true dichotomy, as one entity cannot have contradictory properties at the same time.
Re: (Score:2)
Something that has the property of being grey does not have both the property of being white and the property of being black. Similarly, something that has the property of being a particle-wave does not have both the property of being a wave and the property of being a particle. That we call it a particle-wave is a contingent artifact of our developmental history; on the macro-level, there are no things that behave like particle-waves, but there are things that behave like waves and things that behave like
Is it me or... (Score:3)
...has the "simulated universe" hypotesis just got a slight boost from this finding?
Re: (Score:2)
I think that kid of depends on whether you think that an algorithm that makes something work requires that the universe within which that algorithm appears to be at work, has to be a simulated universe. Is it possible as an alternative that there are several possible processes where these results, or results statistically insignificantly different, might obtain, and it happens that this process wins because it simply uses less energy and produces results that provide better survivability than the other proc
Re: (Score:2)
...has the "simulated universe" hypotesis just got a slight boost from this finding?
No, it hasn't. You would expect things to be governed by simple, predictable principles. Otherwise they'd happen wildly differently every time, and we wouldn't have life at all.
Re: (Score:2)
The simulated universe hypothesis is based on the seeming odd coincidence that our universe’s operation looks identical to information theory.
The problem with that hypothesis is that people seem to forget that our concept of information theory is a function of the universe it was developed in. Thus, it’s no coincidence, and the congruence of physics and information theory is not evidence of simulation.
Re: (Score:2)
I don't think it needs to look similar to the universe's inner mechanisms just because of that.
Re: (Score:2)
Or in the "Red Queen."
So much time would be saved, and so much more understanding of evolution would be had, if sexual selection was thought in schools. My guess is that this isn't done be because of the word sex in the name.
In the context of the article:
survival of the fittest -> narrows the gene pool
sexual selection -> increases variation in the gene pool
The fist part prunes the "bad" genes. The sexual part actually encourages any "bad" genes that became sexually attractive by any random start.
The
How does this differ from John Holland's work ? (Score:3)
Only having read the abstract, and the linked article, I don't really see how this is different from the "2 Armed Bandit" theory which John Holland Laid out 40 years ago in "Evolution in Natural and Artificial Systems". Holland laid out how the combination of sexual reproduction with mutation within a population otpimises search across the space by combining exploitation of good areas of the search space with exploration to find better areas.
Can someone more up to date enlighten me?
kind regards
tree frog
Re: (Score:1)
It's worth reading the paper: http://www.pnas.org/content/ea... [pnas.org]
The point is that the well known properties of the game theory algorithm explain why sex has the effective properties which have been observed but not explained. I haven't read John Holland's paper. Does it refer to the details of the multiplicative weight updates algorithm?
Re: (Score:2)
Not new. (Score:3)
It has been known for years, probably decades, that gene frequencies follow this mathematical rule, and that it has been mathematically proven optimal for solving Multi-armed bandit [wikipedia.org] type problems. Each generation genes are tested by natural selection, and increase or decrease in frequency according to multiplicative increase or decrease. This is a mathematically optimal strategy for exploring and optimizing payoff in a complex unknown environment. Mutation creates random stuff to try, and this mathematically selection algorithm optimally crafts it into useful new information.
-
Re: (Score:1)
And each generation rediscovers for itself that they can publish a paper with this "exciting result!" that is decades old.
This happens a lot when the professor, and the grad students, have never really studied the history of related fields.
Cheap Mathematical Headlines (Score:2)
Why does Slashdot seem to buy in so often to spinning the recurrence of mathematical tools across various fields as some kind of scientific breakthrough? Correlation is not causation, not all structural similarities imply some kind of necessary physical theoretical account. We as empirical agents use logical tools for the formation, quantification and application of theories - so of course some functions will occur in several different settings, because we're bringing the same resources to the table each t
This can't possibly be! (Score:1)
The Earth was created 7000 years ago by some unknown but benign power, evolution doesn't exist - how can some equation possibly describe it using mathematics? Mathematics itself is obviously a made up tool anyway, so you can make it say whatever you want. They'll be using Empirical Science next to debunk what is clearly the One Truth. :)
When I saw the title... (Score:3)
... I imagined that someone discovered a mechanism in genes that favored survival if there were exactly two or three neighboring genes, and non-survival if there were fewer or greater numbers. Oh, and something about a new gene being 'created' when there are exactly 3 neighboring ones.
Re: (Score:1)
Novel, it is not (Score:4, Interesting)
John Maynard Smith introduced the game theory to evolutionary biology in the early 70's. It was a breakthrough at that time, however today it is scarcely news. Evolutionary biology, and in especially population genetics has been a highly mathematized discipline ever since before WWII, when it was developed by Fisher, Wright and Haldane. Later you had Hamilton and Maynard Smith. It is nice that computer scientists noticed that something exciting is going on here, but don't fall for press releases and insubstantiated claims.
Re: (Score:2)
so it has even less credibility then.
Actual Algorithm (Score:1)
Preprint (Score:2)
There's a preprint of what seems to be a more complete paper on the work hosted on arxiv. There's a bit more math in it, but it's still somewhat understandable: http://arxiv.org/pdf/1208.3160.pdf [arxiv.org].