The Simulation Theory and the dangers of pop-science

[PeroK]

I am immediately led to infinite regression in any of these arguments. Is there a rejoinder to "turtles all the way down" or perhaps "simulations all the way down"?

You mean that if and when we start simulating that will be a simulation within a simulation. And we can wait and watch until our simulacra start simulating!

 

[DaveC426913]

I think I agree that the alternatives are outlandish. But I think in a way they seem outlandish because they're so counterintuitive... Remember when QM was coming to fruition and even Einstein didn't really believe it.
I feel as if the simulation idea is more comparable to our current experience, as in we see and use computers every day they are far from alien.
People can more easily visualise this idea and is therefore more easily discussed.
Its like hidden variable is QM people don't like how weird it is so they classical it up a bit by making it deterministic (I've got nothing against HV though)

I think the simulation idea is lazy, and unnecessarily anthrocentric...

Again, I just think most physicists think the same. Its just a lazy bit of guesswork...
Very interesting none the less!

This is exactly what Conway's cells argued, to prove how preposterous it was that they are part of a simulation.

 

[hutchphd]

Right before they died of loneliness...never knew what hit 'em.
If one guesses that there is an infinite reqression of simulacra within simulacra, can one argue as to complexity increase or decease?

 

[JamieSalaor]

This is exactly what Conway's cells argued, to prove how preposterous it was that they are part of a simulation.

Lol good reference!

 

[Chris Miller]

It doesn't really explain the universe/existence any more than the creationist's God does. It just places the answer further out of sight by hypothesizing an unexaminable context. Nonetheless, it is attractive if only in the sense that whatever it all is is playing on the computer that is our mind, not directly observable. The term "simulation" is probably a misnomer, an over simplification or metaphor for a reality we don't have the ability to conceptualize or articulate, even mathematically (yet).

 

[AlexCaledin]

- ah, that's all explained in the Bible. In the beginning, God created the visual-programmable Heaven-and-Earth System and then had seen it to be simulating/producing the good Nature. But Adam chose to hack the System and got completely engaged in its "underground" good/evil Process so now we have to install the Crucial Update to emerge back at the Good output.

 

[Jarvis323]

If one guesses that there is an infinite reqression of simulacra within simulacra, can one argue as to complexity increase or decease?

My feeling is that very deep levels of simulations within simulations (in the way some are assuming) would be very unlikely (impossible?) to be supported by classical computation, due to complexity increases. But we can't easily prove it ($P \neq NP$ would be a start), and quantum computing is a little different, and the universes could be large and old. Scott Aaronson has commented on it,

It seems to me that, to get from here to there, you’d need to overcome four huge difficulties, anyone of which would be fatal by itself, and which are logically independent of each other.

1. As a computer scientist, one thing that leapt out at me, is that Ringel and Kovrizhin’s paper is fundamentally about computational complexity—specifically, about which quantum systems can and can’t be simulated in polynomial time on a classical computer—yet it’s entirely innocent of the language and tools of complexity theory. There’s no BQP, no QMA, no reduction-based hardness argument anywhere in sight, and no clearly-formulated request for one either. Instead, everything is phrased in terms of the failure of one specific algorithmic framework (namely QMC)—and within that framework, only “local” transformations of the physical degrees of freedom are considered, not nonlocal ones that could still be accessible to polynomial-time algorithms. Of course, one does whatever one needs to do to get a result.
   To their credit, the authors do seem aware that a language for discussing all possible efficient algorithms exists. They write, for example, of a “common understanding related to computational complexity classes” that some quantum systems are hard to simulate, and specifically of the existence of systems that support universal quantum computation. So rather than criticize the authors for this limitation of their work, I view their paper as a welcome invitation for closer collaboration between the quantum complexity theory and quantum Monte Carlo communities, which approach many of the same questions from extremely different angles. As official ambassador between the two communities, I nominate Matt Hastings.
2. OK, but even if the paper did address computational complexity head-on, about the most it could’ve said is that computer scientists generally believe that BPP≠BQP (i.e., that quantum computers can solve more decision problems in polynomial time than classical probabilistic ones); and that such separations are provable in the query complexity and communication complexity worlds; and that at any rate, quantum computers can solve exact sampling problems that are classically hard unless the polynomial hierarchy collapses (as pointed out in the BosonSampling paper, and independently by Bremner, Jozsa, Shepherd). Alas, until someone proves P≠PSPACE, there’s no hope for an unconditional proof that quantum computers can’t be efficiently simulated by classical ones.
   (Incidentally, the paper comments, “Establishing an obstruction to a classical simulation is a rather ill-defined task.” I beg to differ: it’s not ill-defined; it’s just ridiculously hard!)
3. OK, but suppose it were proved that BPP≠BQP—and for good measure, suppose it were also experimentally demonstrated that scalable quantum computing is possible in our universe. Even then, one still wouldn’t by any stretch have ruled out that the universe was a computer simulation! For as many of the people who emailed me asked themselves (but as the popular articles did not), why not just imagine that the universe is being simulated on a quantum computer? Like, duh?
4. Finally: even if, for some reason, we disallowed using a quantum computer to simulate the universe, that still wouldn’t rule out the simulation hypothesis. For why couldn’t God, using Her classical computer, spend a trillion years to simulate one second as subjectively perceived by us? After all, what is exponential time to She for whom all eternity is but an eyeblink?

https://www.scottaaronson.com/blog/?p=3482

This was in response to this article claiming to have proven increase in time complexity disproves the simulation hypothesis, which Scott is refuting. Scott thinks the simulation hypothesis is un-falsifiable. It might be important to carefully define falsifiable though. We could probably find strong evidence to suggest that a particular variant of the hypothesis is false under certain assumptions.

Abstract
It is believed that not all quantum systems can be simulated efficiently using classical computational resources. This notion is supported by the fact that it is not known how to express the partition function in a sign-free manner in quantum Monte Carlo (QMC) simulations for a large number of important problems. The answer to the question—whether there is a fundamental obstruction to such a sign-free representation in generic quantum systems—remains unclear. Focusing on systems with bosonic degrees of freedom, we show that quantized gravitational responses appear as obstructions to local sign-free QMC. In condensed matter physics settings, these responses, such as thermal Hall conductance, are associated with fractional quantum Hall effects. We show that similar arguments also hold in the case of spontaneously broken time-reversal (TR) symmetry such as in the chiral phase of a perturbed quantum Kagome antiferromagnet. The connection between quantized gravitational responses and the sign problem is also manifested in certain vertex models, where TR symmetry is preserved.

https://advances.sciencemag.org/content/3/9/e1701758

My personal feeling is that, it may be (if we live in a simulation), that quantum information and quantum supremacy could be due to hidden classical information/computation leaking through the simulation into our reality. So it could be that in the core reality, there is no exploiting quantum strangeness to do better than classical computation, and we only perceive that because we don't see the whole picture.

In other words, maybe some level above us has some trick, that we can't imagine really, like we have quantum computing. And maybe the one above that also has some trick. But maybe all of those tricks are just exploiting processing done at the level above, and ultimately all of the information/processing is accounted for classically, or by whatever type of processing is truly supported in the base reality.

Either way, I think that the deeper the level of simulations within simulations, the more you would probably lose, so to speak. Maybe the simulated universes would get smaller and or less accurate or rich, or time will pass much slower. In actuality, if you did buy the probability argument, it may be that the probability of being in level $n$ is exponentially smaller than being in level $n-1$. Even if it's not exponential, but only a constant increase in run time/space at each level, if the initial simulation is finite, then it would either bottom out eventually, or asymptotically, time would be approaching a stand still.

But the base reality could be really really big, or even infinite. And time could be infinite. So what could we say then? Like Scott says,

'For why couldn’t God, using Her classical computer, spend a trillion years to simulate one second as subjectively perceived by us? After all, what is exponential time to She for whom all eternity is but an eyeblink?'

Of course, simulating a universe as rich as yours, but at a much slower rate, would also probably imply that the universe is not all computed simultaneously in parallel. So for example, some parts would be waiting/slowed/frozen while others are updating. We don't have an absolute reference for time, and have relativity of simultaneity. So that seems on the surface sort of consistent. If it were to be the case that we seemed to have absolute simultaneity and time, then maybe that would be evidence against a simulation hypothesis.

It's also interesting to think about how the simulators would plug-in, so to speak, if the simulated reality is exponentially slower than the base reality. Could they somehow experience time differently while in the simulation? It would be pretty boring to plug into, or even watch a simulated reality if it took several years in your time to see 1 second of theirs.

We could also just have a many simulation hypothesis, that doesn't rely on recursion, but just a bunch of universes each with one level of simulation. Or there could be only one core reality and one simulation for all we know.

 

[Stephen Tashi]

In Wigner's Friend type scenarios, one viewpoint of analysis represents the both friend and a physical system that the friend experiences. This representation resembles (or is !) the mind of the analyst running a simulation of the friend and the friend's environment - not a simulation that represents every detail of a realistic situation ( not the color of the friends eye's, the weather outside the lab etc.), but a simulation that represents aspects relevant to the analysis.

So a moderated version of "Everything is a simulation" is that every correct physical description describes a simulation. If we describe a physical situation that involves "an observer", we imagine that observer functioning as if the observer exists within our simulation.

It's not suprising that apparent logical paradoxes can arise if a person (e.g. Wigner) tries to analyze a situation where that same person is represented in the analysis. (People who explain away the logical paradoxes of Wigner's friend usually do so with scenarios that put Wigner, his friend, and their physical environments in the analysis. So the paradox-explainer is not faced with the the problem of putting himself in the scenario.)

 

[Digcoal]

This is the issue I find with the simulation hypothesis. It must start with a good formalization. What is a simulation, and how is it different than a non-simulation? Does it have to be intelligently designed? Self contained? Embedded within another reality? Depending on the formalism, you could argue that many religions are forms of simulation hypothesis.

The other issue I have is with Nick Bostrom's application of probability.

Anyway, as long as a hypothesis begs answering lots of deep mathematical problems, I don't see the problem.

Your, and everybody else’s, perception of “reality” is 100% simulation.

There is a deep irony in a simulated conversation about “Simulation Theory” being a religion.

 

[Digcoal]

Remember the operative word 'simulate'. It's not 'recreate'.

Combine the 'Simulation' conjecture with the 'Brain in a Jar' conjecture and you won't have to simulate atoms. You merely simulate the observation of atoms and feed it to the brain. In other words, all those atoms out there in the universe are not really there, they're a projection on our senses.

Interestingly, this creates the kind of reverse Holographic Principle: that the entire universe is merely a projection on the inside of spherical screen that surrounds us like a hollow shell.

Have you ever considered how similar ‘collision detection’ in 3D game programming is eerily similar to the electromagnetic field?

 

[Ralph Dratman]

This is the issue I find with the simulation hypothesis. It must start with a good formalization. What is a simulation, and how is it different than a non-simulation? Does it have to be intelligently designed? Self contained? Embedded within another reality? Depending on the formalism, you could argue that many religions are forms of simulation hypothesis.

The other issue I have is with Nick Bostrom's application of probability.

Anyway, as long as a hypothesis begs answering lots of deep mathematical problems, I don't see the problem.

@Jarvis323 Your request for formalization is spot on. "What is a simulation, and how is it different than a non-simulation?"

That is the exact kind of question that needs to be asked in this, as in many "philosophical" problems. For example, long discussions of free will should be prefaced with "what is free will?" --- with the assumption that that initial question has to be decided before any other progress can be contemplated. Often it turns out -- as with "free will -- that there is no satisfactory definition and therefore, I would argue, nothing to discuss.

 

[sophiecentaur]

Your, and everybody else’s, perception of “reality” is 100% simulation.

There is a deep irony in a simulated conversation about “Simulation Theory” being a religion.

I think you may be using a particular definition of "simulation" here. In as far as we use an internal model of all our inputs then that is the only way we an work. Our communications and conscious thoughts are all in the form of analogy and simulation but the 'simulation theory' seems to be looking for some third party (a god figure) who sets up an all embracing simulation and assumes we are all embedded in it, somehow. Very attractive if you want to go as far as possible in the 'concrete' direction
That is all so similar to theism (or deism, whichever) as to be indistinguishable. It will attract people who can't quite come to terms with the idea of a god yet who can't quite commit to atheism.

 

[Digcoal]

I think you may be using a particular definition of "simulation" here. In as far as we use an internal model of all our inputs then that is the only way we an work. Our communications and conscious thoughts are all in the form of analogy and simulation but the 'simulation theory' seems to be looking for some third party (a god figure) who sets up an all embracing simulation and assumes we are all embedded in it, somehow. Very attractive if you want to go as far as possible in the 'concrete' direction
That is all so similar to theism (or deism, whichever) as to be indistinguishable. It will attract people who can't quite come to terms with the idea of a god yet who can't quite commit to atheism.

What definition are you using for “simulation,” and what particular definition do you believe me to be using?

I reviewed my first point for using the G-word when I did not do so in the context that the moderator assumed. I want to be VERY clear that I am not invoking that moderator’s use of the G-word to sanction my previous comment, nor am I invoking ANY notion of the G-word.

If you wish to continue this dialogue with me, I suggest dropping the G-word and any associated concepts to it.

 

[sophiecentaur]

what particular definition do you believe me to be using?

I was assuming that in the "100%" sentence, you were using it in the sense of an internal model or map of our personal world. Clearly we don't have the world actually in our heads so we use a much reduced map. Call it a simulation if you want.
My objection to the 'extended' version of a simulation is that it implies a similar map in the 'mind' of something / someone else. I'm not sure whether you are strongly in favour or against your "G word". I am merely suggesting that the simulation model could be much the same as the theist model - just a bit more at arms length and not involving moral values.
I will nail my colours to the mast and state that I am a committed Atheist. This has nothing to do with Physics and it not a point of discussion. Just like the Simulation model, it ain't falsifiable.

 

[Digcoal]

I was assuming that in the "100%" sentence, you were using it in the sense of an internal model or map of our personal world. Clearly we don't have the world actually in our heads so we use a much reduced map. Call it a simulation if you want.
My objection to the 'extended' version of a simulation is that it implies a similar map in the 'mind' of something / someone else. I'm not sure whether you are strongly in favour or against your "G word". I am merely suggesting that the simulation model could be much the same as the theist model - just a bit more at arms length and not involving moral values.
I will nail my colours to the mast and state that I am a committed Atheist. This has nothing to do with Physics and it not a point of discussion. Just like the Simulation model, it ain't falsifiable.

You seem to be implying that a CPU is a brain and the logical representation of patterns of magnetized particles is its ‘mind.’

I wouldn’t call a CPU a G-word. Would you?

 

[sophiecentaur]

I wouldn’t call a CPU a G-word. Would you?

Of course not but many people seem to need a G word and a massive CPU could be treated, by many people, as one. Polytheism treats pretty much everything as a potential G word.

You seem to be implying that a CPU is a brain

I'd invert that and say that a brain can be regarded as a CPU (unspecified spec). Nervous systems of all levels exist and have a lot in common. If a nematode worm can be characterised in terms of logical functions then why not our own brains?
The philosophical question of a brain analysing itself is a difficult one if one feels there has to be an answer. Personally, I feel no shame in saying that the question is too hard for me (along with a lot of Science and Culture). Reaching for a magic word to take care of all that is one way of dealing with it.

 

[Digcoal]

Of course not but many people seem to need a G word and a massive CPU could be treated, by many people, as one. Polytheism treats pretty much everything as a potential G word.

I'd invert that and say that a brain can be regarded as a CPU (unspecified spec). Nervous systems of all levels exist and have a lot in common. If a nematode worm can be characterised in terms of logical functions then why not our own brains?
The philosophical question of a brain analysing itself is a difficult one if one feels there has to be an answer. Personally, I feel no shame in saying that the question is too hard for me (along with a lot of Science and Culture). Reaching for a magic word to take care of all that is one way of dealing with it.

“Inversion” is just another way of saying two things are analogous, just perceived from different frames of reference.

The point being: mathematics is pure abstraction meant to reduce particular instances to more manageable logical constructions and to reverse the process into other instances of the same abstraction. That’s all the brain ever does. It converts an instance of ‘reality’ into a manageable dataset in order to create a new instance of reality.

 

[sophiecentaur]

“Inversion” is just another way of saying two things are analogous, just perceived from different frames of reference.

Hang on a bit; I inverted the sentence. Draw a Venn diagram of the two statements and the difference is clear. All black objects are not cats.

 

[AlexCaledin]

- anyway, the psychophysical Process of awareness is fundamental, all the rest being the outcome of its self-organizing...

 

[Chris Miller]

I'm a machine whose inputs I call reality. The context/source of this data is unknown. Does a simulation require a simulator?

 

[sophiecentaur]

But isn’t a simulation an artefact, purposely built to represent something? It’s an approximation to reality (or an invented reality). If stands on its own then how is it a simulation? What is it ‘like’?

 

[Nugatory]

This thread has evolved into a convincing example of why the simulation hypothesis is on the list of topics not allowed for discussion here.

It is closed.