Math disproves evolution

[PsychoStick]

I've read these and I'm wondering if this is one of those cases when someone should bitch slap these professors and tell them to stay in their own field of study. I'm just not buying these claims.

http://sensuouscurmudgeon.wordpress.com/2010/01/26/discovery-institute-math-disproves-evolution/

http://www.math.vanderbilt.edu/~schectex/courses/wolfram.html

http://www.unsolvedmysteries.com/usm399463.html

http://ldolphin.org/wmwilliams.html

I've been working off the assumption that evolution is like a story with periods of punctuation. And I don't see how you can apply mathematical probabilities to evolution and say that it is disproved. The odds are phenomenal so it couldn't possibly happen?

Am I just being stubborn or is it actually ridiculous to say math has disproven evolution?

[Angler]

Evolution has already been proved well beyond any reasonable doubt by observation, so anyone who thinks he has used math to disprove it must be using incorrect assumptions in his model. (That's assuming his calculations are correct, which might not be the case).

I believe it was once mathematically "proved" that bumblebees could not fly. Obviously the proof was based on faulty assumptions, since it doesn't jibe with reality.

I don't see how you can apply mathematical probabilities to evolution and say that it is disproved. The odds are phenomenal so it couldn't possibly happen?While I don't have time to read over the material at those links now, you're correct that the matter of evolution isn't a well-defined problem in probability. There is much more than mere random chance at work. It's a lot like trying to calculate the probability that gas molecules in the atmosphere will spontaneously form a tornado. That probability is essentially zero if you treat the problem simply as a mass of randomly moving gas molecules.

[Gorilla]

Better minds than Darwins' may have proven him wrong.

[Macrobius]

I've read these and I'm wondering if this is one of those cases when someone should bitch slap these professors and tell them to stay in their own field of study. I'm just not buying these claims.

http://sensuouscurmudgeon.wordpress.com/2010/01/26/discovery-institute-math-disproves-evolution/

http://www.math.vanderbilt.edu/~schectex/courses/wolfram.html

http://www.unsolvedmysteries.com/usm399463.html

http://ldolphin.org/wmwilliams.html

I've been working off the assumption that evolution is like a story with periods of punctuation. And I don't see how you can apply mathematical probabilities to evolution and say that it is disproved. The odds are phenomenal so it couldn't possibly happen?

Am I just being stubborn or is it actually ridiculous to say math has disproven evolution?

Can you be more specific what you agree or don't agree with? I skimmed all 4 links for formulas or signs of mathematical argumentation, and only the second one had formulas and the last had extensive numerical argumentation but no formulas. Based on this cursory examination, I would be very surprised if the topic has been engaged at all, in any depth, by the sites you ask us to survey and reject.

I also searched for the word 'entropy', which I would expect to occur in a substantive discussion -- it only occurred in the first link, and then not in a context that helped me understand the author's take on it.

I assume you investigated them in depth and have a better grasp of the theories you are rejecting than was offered at the top-level websites. Given my shallow investigation, I would not expect any issues of relevance to be raised at the links. Can you summarise why I should pay attention to them at all? My minimal scan gave me no reason to bother.

This book, available free as a PDF for online viewing, has a nice chapter on the topic (ch. 19): http://www.inference.phy.cam.ac.uk/mackay/itila/ (Direct link to 11 MB PDF: http://www.inference.phy.cam.ac.uk/itprnn/book.pdf [WARNING: 11 MB, PDF])

That should give you a sense of what I would expect to be addressed, how, and with what rigour.

[Impérialiste]

Can you be more specific what you agree or don't agree with? I skimmed all 4 links for formulas or signs of mathematical argumentation, and only the second one had formulas and the last had extensive numerical argumentation but no formulas. Based on this cursory examination, I would be very surprised if the topic has been engaged at all, in any depth, by the sites you ask us to survey and reject.

I also searched for the word 'entropy', which I would expect to occur in a substantive discussion -- it only occurred in the first link, and then not in a context that helped me understand the author's take on it.

I assume you investigated them in depth and have a better grasp of the theories you are rejecting than was offered at the top-level websites. Given my shallow investigation, I would not expect any issues of relevance to be raised at the links. Can you summarise why I should pay attention to them at all? My minimal scan gave me no reason to bother.

This book, available free as a PDF for online viewing, has a nice chapter on the topic (ch. 19): http://www.inference.phy.cam.ac.uk/mackay/itila/ (Direct link to 11 MB PDF: http://www.inference.phy.cam.ac.uk/itprnn/book.pdf [WARNING: 11 MB, PDF])

That should give you a sense of what I would expect to be addressed, how, and with what rigour.

Have you ever heard of the Hardy-Weinberg equilibrium? Hardy is one of the greatest mathematicians of the 20th-century.

[Ahknaton]

This book, available free as a PDF for online viewing, has a nice chapter on the topic (ch. 19): http://www.inference.phy.cam.ac.uk/mackay/itila/ (Direct link to 11 MB PDF: http://www.inference.phy.cam.ac.uk/itprnn/book.pdf [WARNING: 11 MB, PDF])

That should give you a sense of what I would expect to be addressed, how, and with what rigour.
A major flaw with the model presented in that book is that it does not take into account assortive mating.

page 271:

Variation by recombination (or crossover, or sex). Our organisms are
haploid, not diploid. They enjoy sex by recombination. The N individuals
in the population are married into M =N=2 couples, at random,
and each couple has C children { with C =4 children being our standard
assumption, so as to have the population double and halve every
generation, as before. The C children's genotypes are independent given
the parents'. Each child obtains its genotype z by random crossover of
its parents' genotypes, x and y.

[Macrobius]

Yes -- it's just a book on Bayesian Machine Learning and Coding Theory, which happens to have a chapter on genetics 'as an example'. He also gives Race and Crime as an example in another (earlier) chapter -- 3? -- with similar simplicity. I presented it as a baseline expectation of what even the simplest, undergraduate, naive arguments might look like, and what sort of reasoning one should put forth, as a layman (bearing in mind that the title of the thread is 'Math' disproves evolution). It's a very low bar to ask of a discussion, admittedly. My argument is that an 'attack' that doesn't come up to at least that sort of level isn't a serious attempt involving Maths at all -- though there is a level of 'numeracy' common in both liberal and conservative American circles, that is about where the 4th article hits. There should be formulas and calculations in there somewhere, and perhaps a mention of entropy -- I'm not asking a whole lot of the debaters here. Anything less is not out of the gate and definite bunny slope material.

@Imp: As to Hardy-Weinberg, I've heard of it of course. However my knowledge of genetics is minimal. H-W also assumes random mating, I believe.

[Impérialiste]

Yes -- it's just a book on Bayesian Machine Learning and Coding Theory, which happens to have a chapter on genetics 'as an example'. He also gives Racism and Crime as an example in another chapter -- with similar simplicity. I presented it as a baseline expectation of what even the simplest naive arguments might look like, and what sort of reasoning one should put forth, as a layman.

@Imp: As to Hardy-Weinberg, I've heard of it of course. However my knowledge of genetics is minimal.

Sorry Macrobius, I meant to quote PsychoStick. I know you have heard of Hardy.

[Macrobius]

Sorry Macrobius, I meant to quote PsychoStick. I know you have heard of Hardy.

LOL - it's as good a reason as any to link my thread: http://www.thephora.net/forum/showthread.php?t=72567 [Ramanujan was Hardy's prize student, of course]

[PsychoStick]

Can you be more specific what you agree or don't agree with?
How can you use mathematical probability to disprove evolution? I'm just not getting this. How does one prove or disprove the other? This is what I don't agree with.

[PsychoStick]

How can you use mathematical probability to disprove evolution? I'm just not getting this. How does one prove or disprove the other? This is what I don't agree with.

For some reason I'm having a hard time explaining myself.

I don't have a problem with their math. I'm having a problem seeing how math can prove that evolution is a bogus theory.

I believe it was once mathematically "proved" that bumblebees could not fly. Obviously the proof was based on faulty assumptions, since it doesn't jibe with reality.
found this article about that.
http://www.sciencenews.org/view/generic/id/5400/title/Math_Trek__Flight_of_the_Bumblebee
To keep things simple, he assumed a rigid, smooth wing, estimated the bee's weight and wing area, and calculated the lift generated by the wing. Not surprisingly, there was insufficient lift. That was about all he could do at a dinner party. The detailed calculations had to wait. To the biologist, however, the aerodynamicist's initial failure was sufficient evidence of the superiority of nature to mere engineering.

This is the line of thinking that I'm wondering about. Does math actually state that evolution as we know it is impossible, or is it just a statement someone made and then a bunch of people jumped on before it was actually proven one way or the other.

What I'm having the hardest time with is the association between math and evolution. It seems like a flawed argument from the start.
http://www.math.vanderbilt.edu/~schectex/courses/wolfram.html
some of these immensely complicated pictures are generated by very simple rules. For instance, the Mandelbrot set (http://en.wikipedia.org/wiki/Mandelbrot_set), shown at right, consists of the set of points c in the complex plane for which the iteratively defined sequence z0 = 0, zn+1 = zn2 + c does not tend to infinity. Iterative applications of simple rules produce many amazing fractal images, and some of them strongly resemble images seen in nature. An example of this is Barnsley's fern (http://www.home.aone.net.au/byzantium/ferns/fractal.html), shown at left. The works described above are experiments conducted on computers, but the ideas are mathematical. The certainty of mathematics is greater than that of any of the other sciences. In a chemistry or physics experiment, there is always the possibility that the outcome is influenced by some phenomena that have not yet been observed -- e.g., that our instruments aren't sensitive enough, or our perception is not unbiased enough. But a problem in mathematics is finite; a mathematics problem has only the ingredients that are put into the problem by the mathematician. There are no hidden influences, and so we can be completely certain of the outcome.


The advocates of intelligent design are simply wrong when they say that ordered complexity cannot occur by accident. Wolfram's mathematics shows in examples that ordered complexity does occur by accident.

How does a mathematical algorithm take into account the unseen influences of evolution? As he says above, a mathematics problem has only the ingredients that are put into the problem by the mathematician.

So how can he prove or disprove something like evolution with math?

[Macrobius]

Well, I suppose one might as the flip question -- how can maths be used to demonstrate evolution. (The modern theory is inherently quantitative, and more concerned with population counts, and counts as evidence of health, than 'qualitative' health of the organism)

The theory of evolution is (among other things) a theory about the gene frequencies of populations of biological organisms, and how those frequencies change (evolve) with time. The mathematical part is the differential equation that models and predicts the frequencies. If a population is stable (in genetic equilibrium...) then it has reached a fixed point -- the solution to the equation 'reproduces' itself. This is a mathematical ideal that isn't exactly realised -- like the idea of a solid or liquid reaching 'thermodynamic equilibrium'. These properties are statistical properties of large systems -- that is to say, they are mathematical results and not physical results. If one can find a contradiction, then either the model is bad or is being misapplied ('the theory is false'). If you decompose a water molecule and get 3 hydrogens and 2 oxygens, rather than 2 and 1 respectively, then the theory the molecule's structure is H2O is falsified -- by the mathematics of counting. Likewise the mathematics of thermodynamics is just combinatorics -- and the mathematics of evolution is combinatorics.

The mathematics behind evolution is more complex than counting 3 things, but it's still maths, and it is still falsifiable in a purely mathematical (i.e., combinatoric, statistical) way.

It is a funny sort of way to put it 'Math disproves a Physical Theory', but we can parse that 'the physical theory is mathematically inconsistent.' We note these sorts of things all the time -- a model has a singularity that is 'unphysical', or we have an infinity in the calculation of the energy of a system that we 'can't renormalise away.' In those cases, some feature of the mathematics in the model causes it to fail, at least some of the time.

[PsychoStick]

So it doesn't actually prove it wrong, it just calls it mathematically inconsistent with what we would expect?

[Macrobius]

Speaking now to intelligent design -- which I think is the elephant in the room here. The 'goal' of the theory of evolution as a philosophy and anti-religious weapon (as opposed to Science) is to demonstrate that the 'apparently designed' features in nature are not *caused*, but arise as correlations in a predictable, descriptive model.

That is, it may *appear* that organisms have a purpose, and were caused for that purpose, but we can deconstruct this 'causation' into correlations -- a theme in 20the century science, to which the Neo-Darwinian synthesis belongs [1930s].

Here is an instructive and purely mathematical example:

http://en.wikipedia.org/wiki/Simpson%27s_paradox

Simpon's paradox is not a real paradox. It's a property of certain contingency tables [predicted vs. observed gene frequencies would be an example of a contingency table], that they 'look' causative when they are demonstrably not. In fact, picking tables at random, one would expect this property to show up in a minority of cases, but a fair fraction of the time, so it is a possibility in *any* scientific study.

The illusion of paradox arrives because *if* you believe a certain causal model, *and* that causal model happens to be false, then you will be *surprised* by certain correct predictions from the correlations.

The 'play' evolutionists make, is to present the biology of terrestrial organisms as a giant 'simpson's paradox' for theists -- to show their beliefs in causation -- never mind *intelligent* causation -- are falsified, and that teleology is not only unnecessary, but downright misleading, in Science.

A problem that must be squarely faced by such an attempt, however [and this is the substantive point of my post] is that causative reason can *never* proceed from rigourous analysis of correlations. In fact, the conditions of causative reasoning are very interesting to AIs -- we humans reason about causation intuitively. A computer, however, will never be bothered by Simpson's paradox -- and will, by some sort of karmic equality, choose things that appear absurd to us, precisely because we know they are causally impossible.

The axiomatization of a causal calculus, to supplement the logic of correlation, is quite recent:

I strongly urge anyone interested in the mathematics to consider the work of Judea Pearl:

http://bayes.cs.ucla.edu/jp_home.html

[Macrobius]

So it doesn't actually prove it wrong, it just calls it mathematically inconsistent with what we would expect?

Putting it that way raises thorny issues about the role of hypotheses in scientific reasoning [they were scorned in the work of Newton and Maxwell].

As I state in my other response -- the real issue is whether causal inference is mathematically [in a Bayesian sense, say] warranted, and whether the claims being made by *either* side are susceptible to any formalisation that we could call logical, rather than rhetorical, heuristic, polemic, etc.

Mathematical possibility is part of the picture -- maybe the key part -- but I don't think either side has done its homework.

In any event, a mathematical inconsistency would definitely qualify as 'proving a theory wrong'. If not, what sorts of inconsistency, if any, do you accept as the proof of anything whatsoever? Do you accept 'indirect proofs', or are you a strict Intuitionist? If I prove their *had* to be two gunmen, based on scientific evidence, in RFK's assassination, have I used mathematics to disprove Sirhan Sirhan acted alone? Of course I have. Mathematical theories such as thermodynamics or evolution can and do fail in strictly mathematical ways, by yielding up contradictions. The problem is to formulate what those falsification conditions might actually be, for evolution -- much harder.

[Impérialiste]

I suspect the relationship between math and the physical world is purely coincidental and therefore the two are dimensionally different, if you will. For example, what of the physical world is there of the infinitesimally large and the infinitesimally small? These are mathematical abstracts that prove points, and some of these points have physical relevance. But it's a purely coincidental relationship.

Think of determinants and traces. They are mathematically relevant, but no self-respecting physicist would ever use such concepts. Instead, they use approximation methods that cut calculation time in half while extracting the same bit of information.

On its own, I don't think math can prove a lot, if anything, about the physical world. If you have ever studied mathematical models of the physical, they are severe oversimplifications. Indeed, the physical can often be a lot more complicated than what mathematical abstracts would have one believe. Part of the reason, but hardly the whole reason, for this is that mathematical calculations are difficult to carry out, which is why theorems are the heart and soul of pure mathematics: Consider how easy it is to calculate a differential equation like y'' + y = 0 and how much harder (in fact, impossible) it is to calculate y'' - k(1 - y^2)*y' + y = 0. If one lets k = 0, then one has y'' + y = 0, and that is easily solvable. [y(x) = A*cos(x) + B*sin(x).]

[PsychoStick]

Mathematical possibility is part of the picture -- maybe the key part -- but I don't think either side has done its homework.

In any event, a mathematical inconsistency would definitely qualify as 'proving a theory wrong'. If not, what sorts of inconsistency, if any, do you accept as the proof of anything whatsoever?
This is only true so long as the math is correct, and has the correct inputs (sorry, I'm an electronics technician and my brain works in a very straightforward input to output sense. I also tend to think along the line of logic gates also). It's the same as the bumble bee myth. Mathematicians can say look, it doesn't add up. And they are right, but they aren't using the right information. It's like the Zeno paradoxes. The math shows proof of impossibility for something we know to be possible.

A guy named Roger Penrose claims that the probability of an ordered universe just happening is incredibly absurd, but i can't find the math he used to come up with this claim.

I'm starting to read this:
http://www.mathematicsofevolution.com/

But there's a ton to read, i'm starting with the article, and if it isn't complete bullshit I'll read the book.

And thank you to Macrobius for taking the time to explain some of this to me.

[Impérialiste]

This is only true so long as the math is correct, and has the correct inputs (sorry, I'm an electronics technician and my brain works in a very straightforward input to output sense. I also tend to think along the line of logic gates also). It's the same as the bumble bee myth. Mathematicians can say look, it doesn't add up. And they are right, but they aren't using the right information. It's like the Zeno paradoxes. The math shows proof of impossibility for something we know to be possible.

A guy named Roger Penrose claims that the probability of an ordered universe just happening is incredibly absurd, but i can't find the math he used to come up with this claim.

I'm starting to read this:
http://www.mathematicsofevolution.com/

But there's a ton to read, i'm starting with the article, and if it isn't complete bullshit I'll read the book.

And thank you to Macrobius for taking the time to explain some of this to me.

Just to let you know, someone like Roger Penrose would never go against evolution as a scientific theory. In fact, no self-respecting scientist would unless the scientist knew nothing about evolution. Evolution is a fact.

[PsychoStick]

Just to let you know, someone like Roger Penrose would never go against evolution as a scientific theory. In fact, no self-respecting scientist would unless the scientist knew nothing about evolution. Evolution is a fact.

http://www.faizani.com/news/news_2003/math_impossibility.html

[Impérialiste]

http://www.faizani.com/news/news_2003/math_impossibility.html

What's the source of this article? It's posted by a religious site, sites that often claim Albert Einstein believed in some God. Penrose is a known Atheist. I read a lot of his book on physics. He goes against string theory but supports relativity.

[PsychoStick]

What's the source of this article? It's posted by a religious site, sites that often claim Albert Einstein believed in some God. Penrose is a known Atheist. I read a lot of his book on physics. He goes against string theory but supports relativity.

didn't really look at the page did you? Its source is at the bottom.

Roger Penrose, The Emperor's New Mind, 1989; Michael Denton, Nature's Destiny, The New York: The Free Press, 1998, p. 9

[Crowley]

I doubt if life is the result of coincidence, or offbeat chance. I guess that life is inherent when certain temperatures and certain organic arrangements occur, which given the eternally expanding and contracting universe is mathematically certain to recur over and over again -- or, lets say, the universal big bang happens only once, then still when certain temperatures and organic arrangements occur life will emerge spontaneously -- only to die when the suns burn out.

[Ahknaton]

What's the source of this article? It's posted by a religious site, sites that often claim Albert Einstein believed in some God. Penrose is a known Atheist. I read a lot of his book on physics. He goes against string theory but supports relativity.
Penrose is basically a Platonist. That's the context in which his comments about an "the probability of an ordered universe just happening" being extremely small should be taken. In his view the order comes from some reflection within the material universe of the Platonic realm, not from God. He's not pushing intelligent design, but on the other hand he's not a strict naturalist/materialist/nominalist like many Darwinists are.

He also speculates that our minds might exist in a separate mental realm that our brains can mysteriously communicate with via quantum processes.

[Macrobius]

He also speculates that our minds might exist in a separate mental realm that our brains can mysteriously communicate with via quantum processes.

There is good reason for this speculation, known as the 'measurement problem' in quantum mechanics. The problem in 'orthodox' quantum mechanics is that the Schroedinger equation is basically linear and deterministic, but has to be combined in an ad hoc way with the 'Born rule' that states (under various assumptions) the observed values are sharp and non-deterministic -- a discrete probability distribution with weights equal to the square of the wave function's modulus, and values equal to the eigenvalues corresponding to the basis states. Von Neumann showed, or purported to show, that the 'wave function collapse' that gives QM its non-determinism can be located [I]anywhere 'between' the microscopic realm, in the apparatus, or in the mind -- and there is *no testable prediction* that can possibly identify where it happened, at least from within the theory, since that detail is non-observable.

Von Neumann and Wigner then proposed that human consciousness *causes* the wave function collapse -- and they are proof from any refutation, if VN's theorem is correct, since no evidence is *possible* within the framework of the theory that can refute it. Since the interpretation is (on the theory) non-falsifiable, it is not a proper subject for Scientific claims, and may be decided arbitrarily. Furthermore, this arbitrary choice always gives the right answers, relative to QM's validity, so it is, as we say, 'without loss of generality' (or rather, independent as an axiom). As a possible interpretation -- unless some mathematical discovery, such as that recently made in the PBR paper elsewhere discussed -- it will stand exactly as long as QM itself stands as a viable physical theory of how the world works.

Thus, I have on occasion pointed out that this implies QM conflicts with the evidence for Evolution (for example), to the extent that evidence is claimed to establish some observable fact -- since any *experiment* or indeed *observation* of modern science necessarily involves consciousness of humans, and consequently if that consciousness is *causative* of the result (causes the collapse) then such collapses can only have been going on so long as consciousness of the evolutionary sort that 'does that' has been around, and the quantum description is limited to the ability to *communicate* scientific results resulting from quantum collapses, and therefore the universe as described by QM can only be at most as old as the invention of writing -- about 6,000 years, in agreement with the Bible. The other states observed, 'older evidence from fossils', are collapsed from whatever complete set of states described the universe before human consciousness destroyed it ('The Fall') and made it into the single, grossly corporeal 'mixed state' thing we observe today, wherein our mortality results from the -- observed and therefore 'intentionally' caused -- physics that governs our *current* observations.

That is to say, the radical 'epistemic' interpretation of QM is somewhat inimical to Evolution, or at least the (Realist and Objective) way we usually regard the fruits of Science -- namely, as *fact*, as Imperialiste put it. Something about the *objective* world, that is the case.

Penrose gets around consequence by positing a *non-linear* mechanism involving gravity that causes the quantum collapse -- he wants to add terms to the Schroedinger equation anyway, to accommodate gravity, so this costs him little and might get him out of a bind. In other words, the universe obeys the deterministic linear propagation *until* it reaches a certain scale (set by gravity), at which point the collapse mechanism kicks in. This is an 'ontic' view. To fit with von Neumann and the 'epistemic' tradition, he locates this ontic effect in the brain [remember, he is as free as von Neumann to do so]. But he adds detail as to *how* the collapse occurs. Or maybe we should say 'why'.

If you believe the 'orthodox' reductionist account of Science and don't take up with von Neumann and Wigner, you run into a problem -- the brain itself really should be describable by chemistry and physics, and thus quantum mechanics too, but then the collapse cannot occur in the brain, because macroscopic systems like the brain -- which have on the order of 10^23 atomic wave functions times some few more orders of magnitude -- ought to be linear suppositions of the pure atomic or molecular states, and cannot *themselves* be the cause of a collapse (**). Many worlds theory (another ontic approach) gets around this by claiming the states are mostly decohering, so all the 'other possibilities' are just as real, but we can't observe them. Bohm avoids collapse altogether, but at a price of making the wave function holistic and dependent on the state of the entire universe .

(**) My point here is that the 'orthodox' interpretation already has something 'spooky' about it: namely that if measurements are, in fact, made by brains, and brains are described by quantum mechanics, then at the crucial point in the measurement where collapse occurs -- something else besides the brain (the ghost in the machine?) -- must be measuring the measurement, and so on in infinite regress until we reach *something* that measures a measurement of a measurement... that *isn't* a material system obeying QM. If you believe Materialism, you either have to consider QM a refutation of your position, or find a materialist mechanism that yields the measurements -- since those *violate* the Schroedinger equation, and require an arbitrary and non-deteriministic *something* to intrude. The 'orthodox' approach is to prohibit the student from asking 'and what might that be'. It is Just So, and students who ask too many questions are prone to disappear (from academic physics). Of course, if one is allowed such magisterial prohibitions, we might also add in 'and furthermore the student is prohibited from questioning dogmas of the Catholic Church, or examining evidence that Theology might be wrong, or suggesting interpretations which so imply.' Which doesn't sound so 'scientific' after all. Positing the Born Rule and the existence of Quantum Collapse to Randomness, as a dogma, is just a pernicious (or just as necessary?) as any other restriction on scientific inquiry. Thus far, [i]and no further.

Penrose's approach is, then, one of the more 'scientific' out there, if by science you mean 'in accord with our modern, atheistic sensibilities.' The criticism of his choice of interpretation is that this is primarily an aesthetic criterion -- some really cool possible connections no one has proven and may never prove -- not one based, particularly, on evidence.

[Macrobius]

Article at the New York Times about quantum collapse, decoherence, and the practicalities of designing a system capable of simulating a quantum system (you know, like reality...)

http://www.thephora.net/forum/showthread.php?t=78144 (excerpts and article links at this thread)

My own answer is that the brain, as a pulsed neural net (PNN) constructed out of quantum components, is already capable of simulating non-linear dynamical systems arbitrarily (see my TOE thread for the relevant link) -- so we don't need quantum computers since we have brains that can already measure things and simulate reality in a reasonable and obvious way, and predict the outcome of complex quantum problems, such as 'what is that bear about to do, and should I care?'. [The brain's neural net is a little bit 'fractal' in the sense it can run *different* simulations on different time/length scales, simultaneously -- it isn't locked into a single induced interpretation of what it is modelling at a given instant. Also, the read out is somewhat analogue, except for our conceptual symbolic processing.]

Penrose's other point (about zombies and AIs) is that building algorithmic computers out of deterministic gates won't get us to intelligence. Plausibly, a quantum computer might be able to 'simulate itself', but I suspect the classical approach taken by Nature in Evolution is, somehow, more effective and more elegant. ADDED: In fact, a quantum computer may be harder to generate AI from, since it has to face the question of 'what happens if I become conscious and generate my own self-collapse'. At least classical brains appear to have solved *that* problem. Classical brains control the quantum collapses themselves, perhaps as a constitutive feature of intelligence. Quantum brain, like Soviet brain, collapse *you*.

Perhaps when we die, we see a bright flash, then read out a single random integer -- according to Evolution, an encoding of our own genetics.

[PsychoStick]

Penrose's other point (about zombies and AIs) is that building algorithmic computers out of deterministic gates won't get us to intelligence. Plausibly, a quantum computer might be able to 'simulate itself', but I suspect the classical approach taken by Nature in Evolution is, somehow, more effective and more elegant. ADDED: In fact, a quantum computer may be harder to generate AI from, since it has to face the question of 'what happens if I become conscious and generate my own self-collapse'. At least classical brains appear to have solved *that* problem. Classical brains control the quantum collapses themselves, perhaps as a constitutive feature of intelligence. Quantum brain, like Soviet brain, collapse *you*.

Perhaps when we die, we see a bright flash, then read out a single random integer -- according to Evolution, an encoding of our own genetics.

1. quantum computers use deterministic logic gates with the difference being that whereas a bit must be either 0 or 1, a qubit can be 0, 1, or a superposition of both. A single qubit can represent a one, a zero, or, crucially, any quantum superposition of these; moreover, a pair of qubits can be in any quantum superposition of 4 states, and three qubits in any superposition of 8. In general a quantum computer with n qubits can be in an arbitrary superposition of up to 2n different states simultaneously (this compares to a normal computer that can only be in one of these 2n states at any one time). A quantum computer operates by manipulating those qubits with a fixed sequence of quantum logic gates (http://en.wikipedia.org/wiki/Quantum_gate). The sequence of gates to be applied is called a quantum algorithm (http://en.wikipedia.org/wiki/Quantum_algorithm).

2. decoherence on the macro scale isn't an inhibitor in nature. I know 100ms isn't very long, but it's debatable when your working on the quantum level, and it's longer than we have replicated in a lab.
http://blogs.discovermagazine.com/notrocketscience/2010/11/24/how-birds-see-magnetic-fields-%E2%80%93-an-interview-with-klaus-schulten/ you'll have to read the main article also.

3. Quantum computers do not control the collapses themselves. The physical construction of a quantum computer is itself an arrangement of entangled (http://en.wikipedia.org/wiki/Quantum_entanglement) atoms, and the qubit represents both the state memory and the state of entanglement in a system. A quantum computation is performed by initializing (http://en.wikipedia.org/wiki/Initialization_%28programming%29) a system of qubits with a quantum algorithm (http://en.wikipedia.org/wiki/Quantum_algorithm) —"initialization" here referring to some advanced physical process that puts the system into an entangled state.

4. You still run programs on a quantum computer, but now instead of fuzzy logic you can straight write the programs to use the registers that allow for the superposition up/down. And as it stands right now emulating a quantum computer from inside a classical computers OS is a better option.
http://rugth30.phys.rug.nl/qce/

[Macrobius]

1. quantum computers use deterministic logic gates with the difference being that whereas a bit must be either 0 or 1, a qubit can be 0, 1, or a superposition of both. A single qubit can represent a one, a zero, or, crucially, any quantum superposition of these; moreover, a pair of qubits can be in any quantum superposition of 4 states, and three qubits in any superposition of 8. In general a quantum computer with n qubits can be in an arbitrary superposition of up to 2n different states simultaneously (this compares to a normal computer that can only be in one of these 2n states at any one time). A quantum computer operates by manipulating those qubits with a fixed sequence of quantum logic gates (http://en.wikipedia.org/wiki/Quantum_gate). The sequence of gates to be applied is called a quantum algorithm (http://en.wikipedia.org/wiki/Quantum_algorithm).

2. decoherence on the macro scale isn't an inhibitor in nature. I know 100ms isn't very long, but it's debatable when your working on the quantum level, and it's longer than we have replicated in a lab.
http://blogs.discovermagazine.com/notrocketscience/2010/11/24/how-birds-see-magnetic-fields-%E2%80%93-an-interview-with-klaus-schulten/ you'll have to read the main article also.

3. Quantum computers do not control the collapses themselves. The physical construction of a quantum computer is itself an arrangement of entangled (http://en.wikipedia.org/wiki/Quantum_entanglement) atoms, and the qubit represents both the state memory and the state of entanglement in a system. A quantum computation is performed by initializing (http://en.wikipedia.org/wiki/Initialization_%28programming%29) a system of qubits with a quantum algorithm (http://en.wikipedia.org/wiki/Quantum_algorithm) —"initialization" here referring to some advanced physical process that puts the system into an entangled state.

4. You still run programs on a quantum computer, but now instead of fuzzy logic you can straight write the programs to use the registers that allow for the superposition up/down. And as it stands right now emulating a quantum computer from inside a classical computers OS is a better option.
http://rugth30.phys.rug.nl/qce/


Not sure what you are trying to say. Your response uses cut and paste sentences from the wikipedia articles on qubits and the one on quantum computers (neither linked), interspersed with what appears to be original text and elaboration of your own. What point exactly are you addressing? You also link an article about magnetic navigation in birds.

You realise why you can't simulate a quantum algorithms specific to a quantum computer, including factoring or computations in quantum chemistry, on a clssical algorithmic device, in polynomial time right? BQP likely != P. It is only 'the best option' in a research sense, for problems where it is, in fact, irrelevant to be able to do so.

[PsychoStick]

Not sure what you are trying to say. Your response uses cut and paste sentences from the wikipedia articles on qubits and the one on quantum computers (neither linked), interspersed with what appears to be original text and elaboration of your own. What point exactly are you addressing? You also link an article about magnetic navigation in birds.

You realise why you can't simulate a quantum algorithms specific to a quantum computer, including factoring or computations in quantum chemistry, on a clssical algorithmic device, in polynomial time right? BQP likely != P. It is only 'the best option' in a research sense, for problems where it is, in fact, irrelevant to be able to do so.
Point 1. quantum computers still use deterministic logic gates just as classical computers do even if their gates allow added function.

Point 2+3. addresses: to face the question of 'what happens if I become conscious and generate my own self-collapse'. This is only possible with shitty programming. On a quantum computer the basics still apply to some degree. Any program that runs would be sustained in 'ram', the logic functions change because of the quantum register, and runs calculations through a 'processor'.
Any program that becomes self aware isn't going to collapse itself because the actual program runs the algorithms itself through the register while maintaining itself in an entangled state on the 'ram'. Even if collapse (into a single state) occurred because the program was aware of itself it would not cause decoherence of it's own entangled state because the collapse would occur in the register while the program was maintained in 'ram'. It simply would receive inputs of a collapsed state from the register on the algorithms it ran. Which is how it's actually supposed to work. Reversible logic gates allow for recovery of input data, so even if it collapses the initial state is still recoverable.


Point 4. you answered that one yourself. The bird link was showing that entanglement on the macro scale happens now in nature, and it is only a matter of time before we have it perfected in the lab. When I wrote yesterday I hadn't realized that they had set new records in this field recently.


16,384 dimensions

http://physicsworld.com/cws/article/news/45647

Now, Blatt’s group has again broken the entanglement record, this time with 14 qubits, or an equivalent of 16,384 dimensions. “If one wants to calculate the dynamics of such a system, it’s like simulating the bouncing of a ball in 16,384 dimensions,” says Thomas Monz, a member of Blatt’s group. “Such calculations on a classical computer are still possible but, depending on the quantum system under investigation, can take quite a long time to calculate… For current supercomputers, simulations [are limited] to approximately 43 qubits.”

And as you can see, simulation on a classical computer still tops out higher than quantum computers can actually do.

[Macrobius]

And as you can see, simulation on a classical computer still tops out higher than quantum computers can actually do.

You speak as though these devices exist today. The achievement of the physics group you link -- and it is quite an achievement -- is to create a memory cell14 bits wide. The 'gates' and logic of the program connected to that cell don't exist yet and thus the 14 bits can't compute anything, much less a breathless number of dimensions. If the gates and program did exist, then perhaps the calculations envisioned could be done. Right now, there is no working device that can *use* the memory cell they built -- if it would last long enoug to calculate anything.

High end quantum tech today is to build some gates capable of multiplying the number 3 by the number 5 and getting the answer right -- probably -- most of the time. Sure, you can build a simulator that runs on classical computers. That's what the software you linked does. It can't, obviously, do it any better or faster than the underlying platform hosting the emulator. Assuredly, a simulated quantum computer will be able to do better than multiply 3 x 5. Perhaps, 6 x 7. ;) The QCE simulator program can't calculate anything that you can't calculate faster on a classical computer -- probably on your fingers and toes, assuming you know chisenbop.

This tech is maybe 50 - 100 years off. We're talking the difference between Babbage and the ENIAC or Aiken's Mark-1, when I say that.

[PsychoStick]

This tech is maybe 50 - 100 years off. We're talking the difference between Babbage and the ENIAC or Aiken's Mark-1, when I say that.

I know we had the math for a classical computer 100 years before the materials necessary to build it even existed, but with how fast technology evolves today, I would not be surprised to see a working QC in thirty years.

[Gorilla]

Better minds than Darwins' may have proven him wrong.

Not only was Darwin too stupid to have been taken seriously for starters, his insane theory does not explain much, his theories on this subjuct simply too simplistic, with all of natures' design simply explained by his claim of "sexual" and "natural" selection.

Take lions and tigers for an example. Lions have a spiked penis, and so, tend to be preferred to tiger males by tiger females, or so the examples of captivity would have us believe.

For a mind like Darwins, which explains this better? "Natural", or "Sexual" selection?

There seems no real answer.