You are viewing the historical archive of The Philosophy Forum.
For current discussions, visit the live forum.
Go to live forum

The predicting computer

John Frostell January 05, 2017 at 08:11 10975 views 33 comments
Imagine that determinism is true. Everything happens because of something else in a long chain of events, all following laws of nature, known and unknown to our current science.
In that case everything should be traceable back to the start of universe, if there ever was a start. Then the future must also be predictable (if not so then why?).

Now imagine that a super capable computer of some kind can then calculate everything that will happen given current state of everything and all ongoing processes.

It now predicts exactly what you are going to do and what you are going to think. Can you imagine doing that and not having any other option?

I have serious problems with answering yes to that question, even though I can’t rule out determinism completely.

Comments (33)

m-theory January 05, 2017 at 08:21 #44332
Reply to John Frostell
This is a pretty old idea in philosophy.
It even has a name.
It is called Laplace's demon

You will notice there are a number of arguments against the demon and one which should be of interest, even if you reject the others, is the one from David H. Wolpert.
Marchesk January 05, 2017 at 08:43 #44334
How would the computer compute the future of the universe without being the entire universe? And if the entire universe isn't computing ahead such that we know the future, then you can't have a computer inside the universe doing so.
John Frostell January 05, 2017 at 08:53 #44335
Thank you m-theory. I suspected it was an old idea.

If such prediction is impossible however, than I would say let's live in the illusion of a non-deterministic world, regardless of wether we do or not.
Wosret January 05, 2017 at 09:46 #44338
I don't see how it follows that if the future were determined, and necessary that it would therefore be predictable. If things happened in future tenses that never occurred in any sense in a past tense, then how could it be predictable from observation of past things? Like the occurrence of life before anything like that came about? Funnily enough, the inverse doesn't seem impossible either. That if the universe were not deterministic, that it wouldn't be predictable. Say for instance, that God did everything, but gifted to you a schedule for events, like a prophet or something, then it stands to reason that unless deception or incompetence is afoot, then the future would be predictable through the schedule.

See, what's actually necessary to predict the future to a high degree of accuracy is for the past to contain it in every sense. Everything that can possibly happen would have to be predictable from the initial conditions based on like, a platonic landscape of potential forms or some such.
John Frostell January 05, 2017 at 10:10 #44341
Reply to Wosret

It's allready answered above, but I was not talking about observations but rather laws of nature. So if everything is determined from those laws of nature then at least an observer in another universe could calculate the future if our universe is some sort of predictable machine. I don't really see how the past is even necessary to know, if you know all objects, their current directions and all the laws the obey.

If the laws state the the objects given a situation can only end up one way, then that should be enough to predict the future.
Wosret January 05, 2017 at 11:07 #44347
If it isn't about observation then what does it even mean to "predict the future", similarly if it has nothing to do with the past? What is this knowledge of objects and laws about? What does determinism mean if not the future arises from the past? How could the future possibly be predictable without knowing the past?

My core point to take away, is that novelty is a complete myth in such a universe, and the future has to be contained within the past in its entirety, or at least to the extent that it is predictable. It isn't actually new or different, as it was always perceivable, or discernible from any point in the past before its occurrence. Just saying "the laws predict it" or whatever doesn't quite explain how this could be so, or in what form this could take?
Cavacava January 05, 2017 at 11:43 #44349
My core point to take away, is that novelty is a complete myth in such a universe, and the future has to be contained within the past in its entirety, or at least to the extent that it is predictable. It isn't actually new or different, as it was always perceivable, or discernible from any point in the past before its occurrence.


Not sure I follow. Do you mean that novelty is excluded from a determinate universe or that novelty
isn't actually new or different, as it was always perceivable, or discernible from any point in the past before its occurrence.


I think all novelty rests on the past, that nothing is new under the sun, that what is new is a rearrangement of existing concepts. I doubt novelty in the sense of a novel series, something with no antecedent.

Not so sure about determinism, I think it is pretty much unsolvable problem due to its causal nature, the universe has no reason to be reasonable.

TheMadFool January 05, 2017 at 11:53 #44350
Here's what I think...

These so-called laws of nature have a fatal flaw. Which is...the problem of induction.

We cannot be 100% sure that there are any laws of nature. All scientifically analyzed phenomena could be just a probability streak and may alter behavior any time, any place.

In other words, determinism still needs to solve the problem of induction.

What do you all think?
Wosret January 05, 2017 at 11:53 #44351
Reply to Cavacava

No, I attempted to illustrate that being deterministic and being predictable aren't related.
John Frostell January 05, 2017 at 12:09 #44356
Reply to Wosret

I agree with you on the following

* Our knowledge on laws of nature comes from observations of the past or present.

* If we are interested in knowing why things are the way the are at (i.e. how the became the way) current, we must study the past.

But given knowledge about laws of nature (regardless of how we got that knowledge) and given knowledge about the present state, we can predict the future - if laws or nature only gives one possible outcome.

Novelty is just an illusion in such a universe (I would say thats terrible spontaneously), it something that appears to be unpredictable but was bound to happen all the time. Was that your point too?

If there are any laws of nature for everything, then the knowledge about them - all of them - is enough. What preceeded them in another universe does not affect us today and is there for not of interest in order to predict things.
John Frostell January 05, 2017 at 12:14 #44357
Reply to Wosret

Well if something is predictable than it must be determinstic. At least if we want certainty in our predictions.

The other way around I agree with you on. Things can be determinstic without being predictable, but that seems to Labe the case in extreme cases such as Laplaces demon.
Wosret January 05, 2017 at 12:17 #44358
I wasn't so much arguing against it, as pointing out that what is suggested is Platonism, an idealism, a rationalism that asserts an empirical account that doesn't imply it. Empirically the future doesn't appear to be contained in the past actually, empirically, besides in the sense of some regularities that rather than predicting a different thing in the future from the past, predicts that something will remain stable, or self-similar through time.

The sense in which different things that never occurred yet existing in the past must be an ideal one, and not an empirical one. Just think about what it would mean to be an empirical one for a second.
Wosret January 05, 2017 at 12:18 #44359
Reply to John Frostell

What about my God schedule that perfectly predicts the future?
John Frostell January 05, 2017 at 12:43 #44362
Reply to Wosret

if you must obey that schedule then determinism is a fact. If you can do whatever you want with it - it's a prediction that may or may not be true, just someting that claims to be a prediction.
tom January 05, 2017 at 12:45 #44363
Quoting Marchesk
How would the computer computer the future of the universe without being the entire universe? And if the entire universe isn't computing ahead such that we know the future, then you can't have a computer inside the universe doing so.


Surprisingly, according to the laws of physics, simulating the entire visible universe is indeed possible using a universal computer, from within the universe.

The visible universe is thought to contain about 10[sup]123[/sup] bits of information. A rudimentary quantum computer containing only a few hundred qubits vastly outstrips that!


tom January 05, 2017 at 12:48 #44364
Quoting John Frostell
If such prediction is impossible however, than I would say let's live in the illusion of a non-deterministic world, regardless of wether we do or not.


Such a prediction is impossible for false physics under an out-dated conception of computing.
ChrisH January 05, 2017 at 15:38 #44374
Apparently, prediction from within our universe is not possible. This paper explains why - Determinism and the Paradox of Predictability.

from the paper:
determinism implies external predictability, that is, the possibility for an external observer, not part of the universe, to predict, in principle, all future states of the universe. Yet, on the other hand, embedded predictability as the possibility for an embedded subsystem in the universe to make such predictions, does not obtain in a deterministic universe.
tom January 05, 2017 at 17:01 #44393
Reply to ChrisH

Your link doesn't work.

However, whatever you think determinism implies or otherwise, what can and cannot happen in the universe is solely determined by the laws of physics.

I strongly suspect that the paper you have tried to link to deals with classical computation, operating under the laws of classical physics - i.e. an outmoded conception of computation operating under false physics.
ChrisH January 05, 2017 at 17:15 #44396
Reply to tom

Strange. The link worked when I made the post but it now appears to be unavailable. ETA: It seems to work now.

Here's an alternative paper which makes a similar argument - DOES DETERMINISM IMPLY ABSOLUTE PREDICTABILITY?
tom January 05, 2017 at 17:47 #44400
Reply to ChrisH

False physics.

OK, perhaps that's a bit of an overstatement. There exists a very strange interpretation of quantum mechanics, that is compatible with that one-world determinism. It goes by the name "Super-Determinism". As far as I can tell, there are 3 people who hold that view, one being extremely famous.
Marchesk January 05, 2017 at 18:40 #44407
Quoting tom
The visible universe is thought to contain about 10123 bits of information. A rudimentary quantum computer containing only a few hundred qubits vastly outstrips that!


Do you think that a few hundred qubits could completely simulate a black hole? And by that, I mean the object itself, not just the effect on things outside the event horizon.
Barry Etheridge January 05, 2017 at 19:24 #44416
Quoting tom
The visible universe is thought to contain about 10123 bits of information. A rudimentary quantum computer containing only a few hundred qubits vastly outstrips that!


That's ludicrous. The information within your computer cannot not be within the Universe. If it was to make accurate predictions about the Universe such a computer would need to contain every bit of information in the Universe including the information it contains about every bit of information in the Universe including the information it contains about every bit of information in the Universe including the information it contains about ...... infinite capacity in other words. Moreover just one step in any calculation toward predicting the future of the universe would involve a change of state of part of that information requiring a refresh of all of it to make the calculation valid leading to an infinte regression. The precise future of the Universe is therefore, by definition, incalculable by any calculation device within the Universe no matter how sophisticated. Moreover, by implication, it is equally impossible to calculate the exact history of the Universe to date.
tom January 05, 2017 at 22:19 #44454
Quoting Marchesk
Do you think that a few hundred qubits could completely simulate a black hole? And by that, I mean the object itself, not just the effect on things outside the event horizon.


It depends on the size of the black hole. If the black hole is smaller than the visible universe, then yes. More interestingly, a universal computer inside a black hole could simulate the black hole.

Marchesk January 05, 2017 at 22:58 #44462
Quoting tom
It depends on the size of the black hole. If the black hole is smaller than the visible universe, then yes. More interestingly, a universal computer inside a black hole could simulate the black hole.


What does the algorithm for computing a black hole look like? Do we know that a quantum computer can operate as a turing machine, and do we know that black holes are computable?

Let's say the answer to the first is yes, do we know that a simulation running on qubits is more efficient than a classical computer? I know that quantum computers can perform some algorithms much faster, but I'm pretty sure there are plenty of programs that would not benefit. Like say, running Microsoft Office on a quantum computer.

So let's say you have hundreds of qubits available. Does that mean you could run Office 365 as a service for every human being on the planet?
Marchesk January 05, 2017 at 23:05 #44463
Quoting tom
Surprisingly, according to the laws of physics, simulating the entire visible universe is indeed possible using a universal computer, from within the universe.


Back to this. Let's say all of physics is computable with a few hundred qubits, including black holes and Microsoft Office for all sentient beings insid the visible universe. Do we know what the algorithm for consciousnes is, such that our simulated Office users experience the Joys of making yet another thrilling Powerpoint presentation?
andrewk January 05, 2017 at 23:06 #44464
I don't see how having a computer, quantum or otherwise, that can store enormous volumes of information helps in simulating the universe, if the computer is in the universe. Let the info storable by the computer be A and the info in the universe outside the computer be B. Then, as long as B is nonzero, for the computer to simulate the universe it has to store at least volume A+B which, no matter how mind-bogglingly large A is, will be more than the computer can store.
tom January 05, 2017 at 23:30 #44482
Quoting andrewk
I don't see how having a computer, quantum or otherwise, that can store enormous volumes of information helps in simulating the universe, if the computer is in the universe. Let the info storable by the computer be A and the info in the universe outside the computer be B. Then, as long as B is nonzero, for the computer to simulate the universe it has to store at least volume A+B which, no matter how mind-bogglingly large A is, will be more than the computer can store.


Simple question:

How many bits of information are required to fully describe the visible universe?

How many states are available to a quantum computer comprising 1000 qubits?
Marchesk January 06, 2017 at 00:27 #44507
Quoting tom
More interestingly, a universal computer inside a black hole could simulate the black hole.


So a universal computer could compute the result of itself being sucked into a black hole and having contact with the interior (singularity or whatever lies there).

Although, this seems unfair. I'm guessing the idea of simulating precludes the thing doing the simulating, otherwise we have regress and self-referential issues. I might as well ask if the computer can simulate itself itself being introduced to a really strong magnet, or whatever would disrupt a QC.

Perhaps a more fair question would be could a QC compute a black hole used as a computer? I ask, because there are some articles out there about the possibility of black hole computers:

https://aeon.co/essays/is-the-black-hole-at-our-galaxy-s-centre-a-quantum-computer
andrewk January 06, 2017 at 02:44 #44523
Reply to tom If the quantum computer is in the visible universe, why would the answer to the first not be more than the answer to the second? (assuming they are converted to use the same units. As written above, they seem to have different units, one counting bits and the other counting states).
tom January 07, 2017 at 11:20 #44981
Reply to andrewk

The visible universe could hold ~10[sup]124[/sup] bits of information if it were a black hole of the same size. I've seen arguments that the the number will be of the order ~10[sup]90[/sup] for a realistic non-black hole universe.

So, you need at most ~2[sup]412[/sup] bits of information to describe the universe. (I hope I did the change of base correctly i.e. 10^124 ~ 2^412)

A 1000 qubit quantum computer in a maximally entangled state, would require 2[sup]1000[/sup] bits to describe it!

Whatever is going on in a quantum computer, the Bekenstein bound does not apply!
tom January 07, 2017 at 12:02 #44985
Quoting Marchesk
So a universal computer could compute the result of itself being sucked into a black hole and having contact with the interior (singularity or whatever lies there).


We've got a pretty good idea of what it is like to fall into a black hole, so I don't see why you could not in principle do this - any finite system can be simulated to arbitrary accuracy by finite means.

Quoting Marchesk
Although, this seems unfair. I'm guessing the idea of simulating precludes the thing doing the simulating, otherwise we have regress and self-referential issues. I might as well ask if the computer can simulate itself itself being introduced to a really strong magnet, or whatever would disrupt a QC.


A quantum computer has got to be isolated from its environment. This can never be perfectly achieved, but there seems to be ways around the technical issues involving extra qubits and error correction. It's thus perhaps not too much of a stretch to think of the QC as a black box, with which there is no interaction save for setting inputs and reading the output.

Haven't thought this through, but this feature of the QC might solve any perceived problems with regression etc. The QC is actually in its own Hilbert space!
andrewk January 07, 2017 at 23:35 #45115
Reply to tom Maybe I'm missing something, but it seems to me that if the quantum computer can contain 2^1000 bits of information, and it is in the observable (from Earth) universe, then the observable-from-Earth universe contains more than 2^1000 bits of information, and hence cannot be described in 2^412 bits.

What am I missing?
tom January 08, 2017 at 00:58 #45145
Reply to andrewk

It is undoubtedly counter-intuitive, particularly as there is a great deal of work that indicates that for a classical Turing machine, operating under classical physics this could not be possible. I think the culprit is classical physics.

As I mentioned, the Bekenstein bound cannot apply to what is going on in a quantum computer. This implies that the state of a quantum computer, even its output in general, is not an observable.