Can artificial intelligences suffer from mental illness?

[Michael 345]

How does quantum theory let the universe "compute" the n-body problem; what compute means in this context; and how this relates to Church-Turing. If I could begin to understand the outlines I'd have a much better understanding of the computationalist position. I might even become a convert, although I'd hate for that to happen!

While I think I have the geist of what you are attempting to understand about computation and limitations my take would be the Universe does not compute anything

A bag of marbles tipped out will follow the laws of physics

The bag of stuff from the Big Bang followed the laws of physics

Neither event required computation which, I would contend, requires some sort of intervention (? a god) to put / arrange the objects of the Universe just the way (god) wanted them to go and the computation suggested

I would not entertain that idea for obvious reasons

 

[Write4U]

While I think I have the geist of what you are attempting to understand about computation and limitations my take would be the Universe does not compute anything

A bag of marbles tipped out will follow the laws of physics

The bag of stuff from the Big Bang followed the laws of physics

Neither event required computation which, I would contend, requires some sort of intervention (? a god) to put / arrange the objects of the Universe just the way (god) wanted them to go and the computation suggested

I would not entertain that idea for obvious reasons ...

But there are known (computable) facts other than God. The 4 fundamental universal forces are but a few deterministic properties of spacetime. Why else would we be able to use the terms "laws of physics" if they were not universally applicable?

 

[someguy1]

Neither event required computation which, I would contend, requires some sort of intervention (? a god) to put / arrange the objects of the Universe just the way (god) wanted them to go and the computation suggested

Oh yes perfectly sensible. You may be surprised (or not) that I don't even disagree with you.

I'm implicitly assuming some form of materialism. Everything is made of stuff. We reject Cartesian dualism, spiritualism, any kind of cause that is not explained by physics. There's no non-physical entity designing the system or directing traffic.

I do agree (as many advocates of scientism don't) that materialism is every bit as much a metaphysical stance as dualism. We live in an age of rationality so we think the universe is rational. But we don't know. I don't claim materialism is true. I just take it as an assumption in the rest of my thinking. The game is to explain the world without dualism.

I never imagine that I personally know the ultimate nature of the world. I'm actually puzzled that so many people think they do. That delusion is quite common both on the spiritual and the scientific side.

 

[Write4U]

The game is to explain the world without dualism.

Hazen mentioned this problem. Is the appearance of Life a certainty or is it pure accident? He argues that it is neither because between certainty and pure chance lies a whole range of "probabilities". In the case of Earth, possessing the necessary ingredients of bio-chemistry, it was just a matter of time for life to form.

IMO, the term "dualism" is somewhat misleading in that it assumes two extremes, without considering the range of "probabilities" when talking about large spaces and time frames.

 

[someguy1]

IMO, the term "dualism" is somewhat misleading in that it assumes two extremes, without considering the range of "probabilities" when talking about large spaces and time frames.

I'm not up on the philosophy. I know there are a lot of subtypes of dualism these days. I got lost around substance and property dualism, and the subject's a lot more complicated than that. I'm just rejecting supernatural causes. Of course then I have to define natural and I'm down that rabbit hole. Back to the paper, which turns out not to be 183 pages, but only pages 154-183 in a journal. Much more hopeful now!

 

[Michael 345]

Can the future be predicted, given a powerful enough computer, if if if we know EVERYTHING literally EVERYTHING about the Universe at this current NOW?

I contend NO. The physical aspects of the Universe might be possible to predict for some period into the future but I would not hazard a guess beyond a couple of hundred years and I have reservations going that far

Any predictions become less certain, not more certain, the deeper you go into the future

The biggest stumbling block to accurately predicting the future is the unpredictability of the brain. Mostly the human brain but animal brains are not outside the mix of unpredictable agents which can an do affect the future

When I left the house to shop this morning I didn't know if I would buy Cornflakes or Wheetbix for breakfast. Finished up I bought neither

No computer can present that because I didn't know myself and any choice I might make is subject to change depending on new input I know nothing about before I left the house and encountered this new information

 

[river]

Can the future be predicted, given a powerful enough computer, if if if we know EVERYTHING literally EVERYTHING about the Universe at this current NOW?

I contend NO. The physical aspects of the Universe might be possible to predict for some period into the future but I would not hazard a guess beyond a couple of hundred years and I have reservations going that far

Any predictions become less certain, not more certain, the deeper you go into the future

The biggest stumbling block to accurately predicting the future is the unpredictability of the brain. Mostly the human brain but animal brains are not outside the mix of unpredictable agents which can an do affect the future

When I left the house to shop this morning I didn't know if I would buy Cornflakes or Wheetbix for breakfast. Finished up I bought neither

No computer can present that because I didn't know myself and any choice I might make is subject to change depending on new input I know nothing about before I left the house and encountered this new information

I agree NO , because the Universe is in constant flux . And so for infinity .

 

[Write4U]

I agree that biological organisms bring a non-predictable aspect to the future. Living things do not act in a purely deterministic manner. Evolution of living things itself is a probablistic process.

OTOH, the purely non-organic chemical mathematics could conceivably be calculated out into the future, because of their fixed values.

Does this mean that we could make a distinction of certain things being purely deterministic, while other things are probabilistic? Why should that not be an option?

Here we get back to the concept of "duality", at the extreme ends of a range of probabilities.
But it has been demonstrated that deterministic universal laws do exist for things with specific values and potentials. Their behavior is strictly controlled by their mathematical interactions.

Although there are events, such as super-novae which themselves may be predictable, but in the process create a state of chaos, which may yield a condition which cannot be calculated, except in the most general terms.

But it also has been demonstrated that living things do not evolve in accordance to pure mathematics per se, but are subject to hidden variables, such as mutation, which cannot be predicted either.

I submit that it is impossible to know "everything" now and even less to make any future prediction based on what we do know now.

 

[river]

I agree that biological organisms bring a non-predictable aspect to the future. Living things do not act in a purely deterministic manner. Evolution of living things itself is a probablistic process.

OTOH, the purely non-organic chemical mathematics could conceivably be calculated out into the future, because of their fixed values.

Does this mean that we could make a distinction of certain things being purely deterministic, while other things are probabilistic? Why should that not be an option?

Here we get back to the concept of "duality", at the extreme ends of a range of probabilities.
But it has been demonstrated that deterministic universal laws do exist for things with specific values and potentials. Their behavior is strictly controlled by their mathematical interactions.

Although there are events, such as super-novae which themselves may be predictable, but in the process create a state of chaos, which may yield a condition which cannot be calculated, except in the most general terms.

But it also has been demonstrated that living things do not evolve in accordance to pure mathematics per se, but are subject to hidden variables, such as mutation, which cannot be predicted either.

Hidden variables ? There is nothing hidden about the variables .

 

[Write4U]

Hidden variables ? There is nothing hidden about the variables .

Hidden variable theory
From Wikipedia, the free encyclopedia.
This article is about a class of mechanics theories. For hidden variables in economics, see Latent variable. For other uses, see Hidden variables (disambiguation).

Historically, in physics, hidden variable theories were espoused by some physicists who argued that the state of a physical system, as formulated by quantum mechanics, does not give a complete description for the system; i.e., that quantum mechanics is ultimately incomplete, and that a complete theory would provide descriptive categories to account for all observable behavior and thus avoid any indeterminism.

The existence of indeterminacy for some measurements is a characteristic of prevalent interpretations of quantum mechanics; moreover, bounds for indeterminacy can be expressed in a quantitative form by the Heisenberg uncertainty principle.

Albert Einstein, the most famous proponent of hidden variables, objected to the fundamentally probabilistic nature of quantum mechanics,[1] and famously declared "I am convinced God does not play dice".[2] Einstein, Podolsky, and Rosen argued that "elements of reality" (hidden variables) must be added to quantum mechanics to explain entanglement without action at a distance. Later, Bell's theorem suggested that local hidden variables of certain types are impossible, or that they evolve non-locally. A famous non-local theory is De Broglie–Bohm theory.

https://en.wikipedia.org/wiki/Hidden_variable_theory

 

[river]

Hidden variable theory
From Wikipedia, the free encyclopedia.
This article is about a class of mechanics theories. For hidden variables in economics, see Latent variable. For other uses, see Hidden variables (disambiguation).
https://en.wikipedia.org/wiki/Hidden_variable_theory

Quantum write4u becomes macro , meaning quantum becomes only relevant in the microworld .

 

[iceaura]

Since the universe clearly does solve the n-body problem for an unfathomably large (but finite) value of n,

I don't think you can assert that the universe has "solved" the n-body problem. It's certainly not obvious that it has.

Yes! That's exactly my point. Substitute mind for universe and it's the same argument.

You can't substitute mind for universe - you can't even substitute brain for universe.

In particular this phrase made me feel so validated: In Newton’s laws of physics with point masses, we outline a proof that Church’s thesis is false; physics is unsimulable.

That's irrelevant - boomerangs, to mention one, are unsimulable in theory, but simulated in practice. Brains might very well be as simulable as boomerangs. Even with some digital setup and an algorithm running on hardware, you just have to get close enough.

 

[river]

↑

Yes! That's exactly my point. Substitute mind for universe and it's the same argument.

You can't substitute mind for universe - you can't even substitute brain for universe.

Actually its not a substitution but a real possibility .

 

[Write4U]

Quantum write4u becomes macro , meaning quantum is only relevant in the microworld .

Are you implying that the quantum microworld is not causal to the macroworld?

In fact, IMO, if we could calculate a quantum event, we'd have solved the "uncertainty effect", right?

But I am not even talking about quantum or uncertainty. I am talking about the biological macroworld where unpredictable variables may occur at any given time.

Most living organism are not restricted to pure mathematically defined linear actions, they evolve and act in near infinite forms of expression.
Well, except insects perhaps. But then insects appear to have only auto-response motor capabilities.

 

[river]

Are you implying that the quantum microworld is not causal to the macroworld?

In fact, IMO, if we could calculate a quantum event, we'd have solved the "uncertainty effect", right?

But I am not even talking about quantum or uncertainty. I am talking about the biological macroworld where unpredictable variables may occur at any given time.

Most living organism are not restricted to pure mathematically defined linear actions, they evolve and act in near infinite forms of expression.
Well, except insects perhaps. But then insects appear to have only auto-response motor capabilities.

Mathematics forms nothing .

 

[Write4U]

Mathematics forms nothing .

What about a closed curved line at which all points are equidistant from the center.

If you haven't seen this yet: This may surprise you what shapes are a potentially present in mathematics.
https://www.ted.com/talks/roger_antonsen_math_is_the_hidden_secret_to_understanding_the_world

 

[C C]

Can the future be predicted, given a powerful enough computer, if if if we know EVERYTHING literally EVERYTHING about the Universe at this current NOW? I contend NO. The physical aspects of the Universe might be possible to predict for some period into the future but I would not hazard a guess beyond a couple of hundred years and I have reservations going that far. Any predictions become less certain, not more certain, the deeper you go into the future. The biggest stumbling block to accurately predicting the future is the unpredictability of the brain. Mostly the human brain but animal brains are not outside the mix of unpredictable agents which can an do affect the future.

In a deterministic reality truly regulated by global principles that could be fully captured and represented by abstract quantitative description and the latter's performed equivalent as mechanistic operations... This hotline to a magical "big and as fast it needs to be" computer occupying another ontological level... Could thereby calculate the general fate of large or massive scale systems in the universe which humans would have no control over (including the general fate of the cosmos as a whole).

But knowledge about the future concerning specific and local affairs which people did have the capacity to manage, would result in the applicable governments, agencies, and communities preventing from happening those "events which they considered bad". Thus, making inaccurate/false the original claim of total and perfect prediction.

Actually, such an absolute degree of forecasting would also take into account / predict those very modifications via that future-->past knowledge feedback which compelled humans to make those changes. But since those changes themselves would eventually lead to some alternative "undesired events" which people would likewise try to avoid in response to that adjusted knowledge, the prediction process would collapse into an endless loop of prognostic simulation. Iteration after iteration of results nested inside each other as the process sought to obtain a final state of the local future's path which humans would not respond to or tamper with (ever).

 

[someguy1]

I don't think you can assert that the universe has "solved" the n-body problem. It's certainly not obvious that it has.

The universe figures out exactly -- not approximately -- where to put each and every particle, wave, quark, and galaxy ... each and every moment of time. To do that, a program would need to have a list of each object, from quarks to galaxies; and for each, the exact state of its position and momentum and electric charge and so on; and then compute where it's supposed to go next based on its relation to every other object on the list.

That's a hell of a computation. The universe implements that computation. How does it do it?

Either the universe is a TM; or the universe is a new kind of computation that breaks the Church-Turing thesis.

You can't substitute mind for universe - you can't even substitute brain for universe.

It seems to me that the question of whether the universe is a computation is the exact same question as whether the mind is. But even if they're two different questions, everything I'm saying does happen to apply to both situations. But it seems obvious that the questions are logically equivalent. How could the universe be a computation and the brain not? Or the brain a computation and the universe not? Well maybe that direction's possible. Ok I'll concede the point in this narrow technical sense. The brain could be a locally coherent object in an essentially random universe. A Boltzman brain. But Boltzman brains are nihilistic, like brain-in-vat scenarios. It seems clear enough that the universe is a computer if and only if the mind is.

https://en.wikipedia.org/wiki/Boltzmann_brain

Reporting back about the paper I linked earlier, I've skimmed through it a couple of times. Very interesting. Doesn't actually talk about QM. It turns out (according to the author) that although Newtonian physics is not simulable (can not be simulated exactly by a TM), relativity is.

The paper is extremely technical and way beyond my knowledge of physics. But he did make one very insightful point. In Newtonian gravity, g1*g2/r^2, you get arbitrarily large energy as two point-masses get closer. This is sort of the underlying reason that Newtonian physics isn't a TM. There are infinities.

On the other hand, he claims that relativity fixes that problem, and can be implemented by a TM.

If that were actually true, then it's not out of the question for the universe to be a computation. Or at least for the laws of known physics to be, which is the next best thing.

One problem from my point of view is that he does not talk about chaos at all. My argument is: Since we can't predict the stability of the solar system using computers because the accumulated rounding errors eventually lead to huge differences; therefore a TM can't run the solar system. Nor the universe, for the same reason.

Now the paper, in my first couple of flyovers, seems to play fast and loose with real numbers. I have a very clear understanding in my mind of real numbers, bitstrings, Turing machines and all that ... and I somehow feel that the author is either not being clear enough in his thinking, or else in his explaining. I'm really not convinced by this paper. He's not dealing with the rounding errors.

TMs can not implement real numbers. They can only approximate real numbers. This was perfectly well appreciated by Turing in his 1936 paper. And it seems to me that this is a big problem for anyone trying to claim that the laws of physics, which are based on the math of real numbers, can be implemented as a TM. Whether it's relativity or QM doesn't matter, because those theories use mathematical continuity.

I think he addresses this point so I have to go back and trace this thread of reasoning.

 

[Write4U]

That's a hell of a computation. The universe implements that computation. How does it do it?

Mathematical functions.
The Universe is the computer and it's irreducibly big.

 

[someguy1]

Mathematical functions.
The Universe is the computer and it's irreducibly big.

If the universe is a computer, then either:

* It's a Turing machine; or

* It's not a TM, hence the universe falsifies the Church-Turing thesis.

So which is it? It's most definitely an open question. There are good fundamental reasons why the universe can't be a TM. There are papers claiming it is. Nobody knows.