Slashdot Mirror
Subatomic Darwinism
blamanj writes "In the beginning was Darwinism, then there arose Social Darwinism, now physicists are proposing Quantum Darwinism. According to the Nature article: "If, as quantum mechanics says, observing the world tends to change it, how is it that we can agree on anything at all? Why doesn't each person leave a slightly different version of the world for the next person to find?
Because, say the researchers, certain special states of a system are promoted above others by a quantum form of natural selection, which they call quantum darwinism. Information about these states proliferates and gets imprinted on the environment. So observers coming along and looking at the environment in order to get a picture of the world tend to see the same 'preferred' states."."
... religious Darwinism. IOW, beliefs evolve as previous beliefs are shown to be "unfit," i.e. disproven by observation.
... but, um, see, there's all this little stuff you scientists haven't quite figured out yet about the specifics, and sometimes you argue about it, and THAT'S ABSOLUTE PROOF OF THAT GOD EXISTS AND HE WANTS YOU TO DO EXACTLY AS _____ (insert your preferred version of a frequently mistranslated, politically loaded anthology of folktales here) SAYS!"
"The Earth is flat, because this passage from the Bible talks about God stopping the Sun directly overhead!"
"Um, no, actually, it's a sphere, and here's the proof."
"Okay, okay! But the celestial bodies are little lights in the sky, and perfect and unblemished, and the go around the Earth!"
"Um, no, actually, they've got all kinds of flaws and blemishes, and they all go around the Sun, and here's the proof."
"Oh, damn! But the Earth was created a few thousand years ago, as we can determine from Biblical genealogies!"
"Um, no, actually, it's been around for a lot longer than that, and here's the proof."
"Aaargh! But humans were specially created by God in His image, and are absolutely unique!"
"Um, no, actually, we look an awful lot like other apes, and that's really not a coincidence, and here's the proof."
"*whimper* All right, so the Earth is round, and it and all the other lumpy rocks revolve around the Sun, and it's all really old, and humans are a lot like apes
The correlation between ignorance of statistics and using "correlation is not causation" as an argument is close to 1.
Wait, "Interesting"? Okay, uh...I disagree with your disagreement! ::waits for "Insightful" mod:: :P
"You're older than you've ever been, and now you're even older."
So observers coming along and looking at the environment in order to get a picture of the world tend to see the same 'preferred' states.
Techno-speak for "rose-tinted-glasses"?
Seriously though, thinking about it makes your brain hurt: Did the scientists working on this create the necessary state "they preferred" inadvertently in order to discover the state they wanted to see?
What's next... "market darwinism" when the products people buy survive?
/me needs more eggnog
No. "Darwinism" is about replicators, i.e. organisms that reproduce and that compete for resources.
When used for "Social Darwinism", the word implies that societies reproduce and compete for resources. In many ways this is accurate. You could use "darwinism" to describe many kinds of replicating, competing natural systems.
But quantums...? WTF?
Until we have evidence that quantums are actually lifeforms, the word "Darwinism" is simply not valid.
Anyhow, and on a different note, quantum mechanics is easy. Here's Ites' Dummies Guide to Quantum Physics: matter and energy are made of wavelets, a string of energy. Wavelets look like particles when they're compressed by time or distance. Measuring a wavelet changes it. Wavelets do not breed and they do not compete for resources.
The table is not solid because it's an agreed reality. The table is solid because your hand cannot pass through it.
Sig for sale or rent. One previous user. Inquire within.
So observers coming along and looking at the environment in order to get a picture of the world tend to see the same 'preferred' states.
Sounds a lot like Solipsism which is nothing new at all.
Solipsism is the belief that, because we can only verify our own experiences and no-one elses, only the self is real.
or to put that in Layman's terms: "Go away, you don't exist!"
Why bother with GAs. They just special cases of baysian inference. Say you want to find out probable good values for x (optimization, something else). Define p(x) (starting population in GAs) Define p(d|x) (related to goodness function of GAs) sample from p(x|d)~ p(d|x)*p(x) (natural selection). this is information theoretically optimal (H(X|DATA) = H(X) - I(X,DATA) so it isn't surprise that natural selection does something very close to this.
How do they address this:
There's no way to know exactly how similar different people's perception of the same scenes is;
Quantum-level variations resulting from observation and whatever else are not likely to make a noticable difference in these scenes.
The idea that trees are tending to appear the same way because their particles find their way back to the same place after being displaced by observations isn't implausible, but without further establishing the potential for a contrary situation it seems like overkill!
Good thing they "mathematically proved" that they're right, heh heh.
When I took physics, it was pretty clear that quantum effects are negligible at large scale. For instance, I have a wave form as I'm sitting in this chair at my office. I don't really notice my oscillation all that much.
So, for observation of the macroscopic environment it would follow that quantum effects can be ignored. But then again, I'm arguing against quantum physicists from Los Alamos, so maybe they're just explaining why quantum effects can be ignored at a large scale.
Yet it would seem as simple as "observation of the individual particles of the windows at Buckingham Palace are affected by observation, but statistically speaking, each change is just as likely as the next, so at a macroscopic level the odds of a visible change are infinitesimal." Sure, there's a chance that a window could move but it's so unlikely as to never happen during the life of the universe. I RRTFA and don't see this.
My God, it's Full of Source!
OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
Being aware of an electron does not make it change.
That's not quite true. The thing is quantum mechanics is probabilistic... so that electron is, in general, in some superposition state before you "become aware" of it (i.e. measure it). This means there is some chance it's in state A and some chance it's in state B when measured.
But such superposition states have actual physical consequences as well (probability interference, wave-like properties). Anyway, once you "become aware" of that electron, it is in some exact and certain state. Your measuring inherently changes the state.
This leads to insane arguments and discussions, since the electron (well, more likely a photon in this case, so let's call it such) is used to measure Buckingham Palace, then the photon interacts with your retina to create a measurement event then the pulses carry information to your brain.
So when did the "measurement event" really occur? When did the waveform collapse into a fixed state? Why does consciousness or awareness of quantum states seem to play a fundamental role in quantum mechanics? This must be an artifact of the mathematics behind Q. Mech. and not a real physical phenomena, right?
Unfortunately, as far as I know, there are no really good answers forthcoming from the physics community about these issues. Physicists don't really seem to agree nor do they really care what is meant by "measurement" from a philisophical point of view (e.g. "why do you ask this question, foolish student, can't you see the observation operator is applied to the quantum state, thereby collapsing the waveform?").
Don't take quantum mechanics too seriously - like everything else in physics, it's a great model for answering questions in a certain domain, but it says nothing about the answers to questions outside its domain. Advanced theories, like quantum electrodynamics and quantum field theory have more to say about this topic, I'm sure, but likewise surely leave equally large gaps in knowledge.
The term "quantum darwinism" is really an unnecessary buzzword. There is a certain analogy about states which create many records of themselves surviving a robust pointer states where others are "selected against", but the analogy is really pretty limited and not very useful. It's better to stay away from using terms like darwinism for effect. I should note that I didn't see the word "quantum darwinism" in the title or abstract of either of the actual journal articles this news item references. For the lazy, the two papers in question seem to be this preprint and this article from the Nov 26 issue of Physical Review Letters.
This sounds like an interesting result and Zurek is a premenent figure in the field of quantum decoherence, but this looks like the tying up of some (important) details rather than the revolutionary developement the news article makes it out to be. Even as far back as the work of Everett we had an idea of why two observers who compared notes would always agree on the objective facts. In the many worlds interpretation, this comes down to the fact that if observer A measures system S, there will be many different possible results. So there will be many branches of the wavefunction with A observing each possible result. When observer B measures system S, he becomes entangled with S and A, and there are many possible outcomes, but in each branch of the wave function A and B agree on the outcome. Not sure if that clears anything up. :-) If you're talking about purely quantum systems, the same thing happens in the Copenhagen interpretation. The only tricky part is how to think about it when A and B are "classical observers". Still, I haven't read these papers yet and now I'm eager to.
"You call it a new way of thinking; I call it regression to ignorance!" -- Operation Ivy
> Why bother with GAs. They just special cases of baysian inference.
Classify GAs however you want - we bother with them because they are useful in finding solutions.
This is like saying "why bother with the Fourier Series - it's just a special case of the Fourier Transform." Feh.
it's funny, laught
it's a fitting example of darwinism taking place in the religious memesphere, each time the study of nature and reality shows that some silly religiously based idea can't be backed up, well then the idea gets appropriately modified to fit reality.
Religions claiming to have special knowledge eventually get challenged and many get shown out the door. It's a sad testimony for some religions. And upsetting for the stringent by-the-letter followers.
A fit religion will be one that sticks to the basics: faith, how to live with oneself, how to live with the world and other men. It will not made silly unnecessary assumptions about a reality that cares nothing for how we on the thin surface of a small planet see it.
______________________________________________
sigamajig...
>Modern physicists, believing that wavelets acted a particular way under certain observation arrangements, developed much arcane math and explanation as to why they couldn't completely account for the observed data.
Quantum Mechanics is THE most sucessful and accurate theory ever. whereas the astronomers could not account for the data, QM accounts for the data to ridiculous accuracy and the only problem is accepting the interpretation. and that's a problem with humans, not QM.
No, you can't figure out why it falls - you can only use scientific knowledge of gravity to predict how it should fall. And then you can observe whether or not it actually falls that way.
And if it doesn't fall the way you predicted, you need to go find a new theory.
You haven't answered why the value of G is what it is, or even why gravity exists at all. All you've done is predict the behavoir of two or more masses interacting and labelled the interaction "gravity".
"Every time I hear about Schroedinger's Cat, I reach for my gun." -Stephen Hawking
I think there is a mistaken idea out there that observation alone fixes certain information about sub-atomic particles while the particle is in several states simultaneously beforehand. The problem is with dependence on observation for collapsing the wave form. If this were true, the only cause was observation, then what would count for an observation? How intelligent would the thing causing the collapse have to be to be considered an observer? Can an ape collpase a wave-form? How about a cat? Does it have to be a living thing at all?
Yes the particle holds many states simultaneously at certain times, yes it becomes fixed when an action happens from which an observer can deduce the state in question (whether or not said observer is there to witness it), but no, an actual observer doesn't have to be there to see it at the time for the wave-form to collapse. That's what's ridiculous about Schroedinger's Cat. Once the quantum event which may or may not kill the cat happens, the sucker is alive or dead - period. We just don't happen to know which until we look.
I'm no physicist but that's the impression I get from what I've read. Anyone want to comment?
Happy people make bad consumers.
Therefore, I am God, and it is proven.
Sometimes it's best to just let stupid people be stupid.
There are a host of comments to the effect that Quantum Physics is ad hoc a la Ptolemaic epicycles, or that the research described is pseudo-science etc. First of all Quantum Physics is not ad hoc, nor does it have any relationship to Ptolemaic epicycles. It is grounded in well established axioms and which have proved themselves spectacularly successful in describing physical reality up to and including the physics of the semiconductor devices which commenters used to demonstrate their astounding ignorance and pride therein.
The problem modern physicists face is that the mathematics of Quantum Physics does not obviously lend itself to description in terms of everyday experience. Most people do not have every day experience with superpositions of states nor do they navigate their existence using that model which leads to a disconnect between the mathematics of Quantum Physics and "common sense". That doesn't make the math wrong, it merely indicates that we have adapted to living in a world in which quantum effects can be safely ignored, unless one is trying to make 0.6 micron scale transistors for Slashdotters to abuse.
The research in question actually goes a long way to explaining why it's OK to ignore the quantum nature of reality above certain scales. In short, among the states that a large ensemble of subatomic particles, like Buckingham Palace, can be in there are states which are relatively resistant to large perturbations by observation. Fortunately for the occupants of Buckingham Palace those states tend to describe a palace comoving with the Earth's surface in London, England, and not a palace hurtling towards the sun at a significant fraction of the speed of light. This is a brutally oversimplified plain English explanation of the results, which can only be precisely stated mathematically, and thus likely to lead to significant misunderstanding. Ironically, the research goes a long way to explaining why another reader and I can both agree on the form of the letters of this message.
Is the glass half-full or half-empty?
God, 1200 AD: "Big guy created the whole thing 5200 years ago."
God, 1800 AD: "Clever big guy created the whole thing 5800 years ago. And had to plunk some planets and set up an inverse square law for gravitation. And bury a bunch of weird lizard fossils to confuse us. Either that, or he's been doing some really weird tricks with biology that we're only beginning to guess at."
God, 1950 AD: "Really clever guy (way cleverer than us) created the whole thing out of, umm, something, we don't really know when, but it was a hell of a long time ago, and made particles that behaved like, umm, waves. It's weird and violates common sense, but we can use the math to make televisions. And BTW, now we know how the Sun works."
God, 2004 AD: "Supremely clever dude, existing completely outside of what we perceive as spacetime, may have tweaked an m-brane collision (the math for which only a few hundred of us on the planet can even begin to understand) that resulted in the setting of a few universal constants for the physics engine and the creation of a little bubble of spacetime. Sat back and watched the resulting fireworks for 13.8 billion years to see if sentient life would evolve in a little pocket of it and recognize Him."
Without taking a position either way on the existence or non-existence of God, I humbly submit that the more science we do, the smarter the "God of the Gaps" has to be.
One minor nit-picky point: The "particle" does not "hold many states simultaneously." It is in one state. period. That state is some linear combination (weighted sum) of various quantum eigenstates, but it is only in one state at any time.
When you say "it becomes fixed," you are referring to the collapse of the wavefunction when it is observed. What it collapses to is one of these quantum eigenstates.
You've got a rather basic understanding of the gist of it, but there are some details which you had wrong.
SIGSEGV caught, terminating
wait... not that kind of sig.
No, I'm saying that a Supreme Being won't be able to guess what choice you will make.
Then he's not an omniscient Supreme Being.
My God, it's Full of Source!
OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
GA's are used to maximize arbitrary functions by a mixture of random mutation and crossover between the solution candidates with better aptitude. It's hot stuff, and it comes up with good solutions for analytically untractable problems.
The hard part is usually coming up with a good "gene string" representation. For some things it is fairly simple, but not others.
The simplest I have seen are robot "path" indicator strings. For example, "LRFBRRRL..." Where L=left, R=right, F=forward, and B=backward by say one foot. The problem is that the work-space (simulated or real) generally has to be tolerant to errors such that some reward is still given for partial failures. If the entire string has to be good before any score is acheived, such as the robot having to reach the final goal or it gets zero, then GA won't work very well.
Table-ized A.I.
The observe effect at the micro level does not translate to the macro level being open to observer determination. The odd way the quantum level operates is used to justify all matter of nutty beliefs and ideas. Those who responded with some darwinian (at the quantum level yet) justification bordering on "consensus reality" should have their credentials as scientists lifted.
Its not so much that there must be an observer, but simply that the act of observing (which in order to observe a particle you must either capture that particle or bounce something off of it) changes it. Which in reality makes a lot of sense. It still exist without said observer.