WrongBot comments on Book Review: The Root of Thought - Less Wrong

47 Post author: Yvain 22 July 2010 08:58AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Article Navigation

Sort By: Best

You are viewing a single comment's thread. Show more comments above.

Comment author: WrongBot 23 July 2010 03:20:19AM * 3 points [-]

There is a reasonable argument to be made that "the way it feels from the inside" is just as fundamental as the basic physics of how the world works.

Well, what is it, then?

The problem is that most philosophers who care about phenomenology at all would assign at least some ontologically foundational status to subjective experience, simply because it is foundational enough to you and anyone else with subjective experience.

Ahhhh, I see now. Subjective experience must be ontologically foundational because it feels foundational, subjectively. This seems oddly... circular.

It does mean that Occam's razor should apply to "the way it feels from the inside", which tends to weigh against complex explanations like "configurations of neurons" and in favor of either exotic physics or a spooky superintelligence who can figure out how to run debugger sessions on our physical brains.

Configurations of neurons are not complex. They are complicated, but they can still be explained by the same physics as everything else in the world. You are proposing a more complex universe. Or possibly a god. They are equally implausible without supporting evidence.

Comment author: bogus comment score below threshold [+] (5 children)

Comment author: JoshuaZ 23 July 2010 12:38:01PM * 2 points [-]

You are proposing a more complex universe.

Not really, since both subjective experience and quantum mechanics are part of our universe already.

No, you are proposing a more complicated universe. Quantum mechanical systems can be simulated on a classical computer given a source of randomness. The only caveat, is that if certain compsci conjectures are true then it actually takes more time or more memory for a classical system to simulate these runs than a quantum system would. If the complexity hierarchy exhibits partial collapse with say BQP being equal to P, then even this would in some sense not be true and we'd then have quantum computers as just classical machines with a source of random bits. Now, most comp sci people don't believe that, but the thrust of this argument just requires the fact that classical machines with randomness can simulate quantum machines given extra time and space. since that is the case, in order to assert that quantum mechanics has any chance of causing things like qualia and consciousness would require that there are fundamental gaps in our understanding of quantum mechanics. It would also likely violate many forms of the Church-Turing thesis. So you'd have to basic failings in our understanding of QM and theoretical comp sci for this sort of approach to be even have a chance at working.

Comment author: bogus 23 July 2010 12:53:26PM * -1 points [-]

Quantum mechanical systems can be simulated on a classical computer given a source of randomness.

This implies that unconscious classical systems can simulate a conscious being. But such a simulation of consciousness would not involve the systems in our physical world which can actually be "felt from the inside". In this theory, qualia and consciousness are not caused by quantum mechanics; they are what some extremely complex quantum states feel like.

The only caveat, is that if certain compsci conjectures are true then it actually takes more time or more memory

If quantum algorithms are at all useful, this is enough for evolution to favor quantum computation over classical.

Comment author: JoshuaZ 23 July 2010 01:05:08PM * 3 points [-]

This implies that unconscious classical systems can simulate a conscious being. But such a simulation of consciousness would not involve the systems in our physical world which can actually be "felt from the inside". Qualia and consciousness are not caused by quantum mechanics, they are what some extremely complex quantum states feel like.

At this point how is this claim any different than claiming that these are classical systems and that qualia and consciousness are what those algorithms feel like?

If quantum algorithms are at all useful, this is enough for evolution to favor quantum computation over classical.

That's actually the best argument I've heard for supposing that there's a quantum mechanical aspect to our processing. Thank you for bringing it to my attention. It does make a QM aspect more plausible. However, it is still a very weak argument since a) evolution would only do this if it had an easy way of keeping things in coherence that didn't take up too much resources b) It seems unlikely that there's a substantive evolutionary advantage to any form of computational speedup to processes which we needed to do in the wild. I don't think for example that humans needed to factor large integers in our hunter gatherer societies. This does lead to the idea of deliberately evolving beings that actually use quantum mechanics in their thought processes by selecting for ones that are good at algorithms that do have speedups in a QM system.

Comment author: bogus 23 July 2010 01:41:35PM * -2 points [-]

At this point how is this claim any different than claiming that these are classical systems and that qualia and consciousness are what those algorithms feel like?

Quantum systems have much nicer properties from this point of view. An internally entangled quantum state can be an ontologically basic entity while still possessing a rich internal structure, in a way that has no direct equivalents in classical physics.

evolution would only do this if it had an easy way of keeping things in coherence that didn't take up too much resources

Models of quantum computation are quite variable in how resistant they are to decoherence. Topological quantum computing is much more resistant to errors than models based on ordinary quantum particles.

If there's a substantive evolutionary advantage to any form of computational speedup to processes which we needed to do in the wild.

Why wouldn't there be? Intelligent processing clearly confers some evolutionary advantage, and there have been many proposals for artificial general intelligence (AGI) using quantum computation.

Comment author: JoshuaZ 23 July 2010 01:54:22PM 2 points [-]

Quantum systems have much nicer properties from this point of view. An internally entangled quntum state can be an ontologically basic entity while still possessing a rich internal structure, in a way that has no direct equivalents in classical physics

That makes some sense, although I don't see why a classical simulation of the same wouldn't feel identical.

Models of quantum computation are quite variable in how resistant they are to decoherence. Topological quantum computing is much more resistant to errors than models based on ordinary quantum particles.

This may be true in the same sense that sending a probe to Betelgeuse is easier than sending a probe to the Andromeda galaxy. You are still talking about fantastically difficult things to keep in coherence. We're still talking about systems kept below at most 5 kelvin or so (being generous). It is noteworthy that so far we've actually had far more success implementing standard quantum computers than we have with topological quantum computers.

Why wouldn't there be? Intelligent processing clearly confers some evolutionary advantage, and there have been many proposals for artificial general intelligence (AGI) using quantum computation.

There's no evidence of any process we associate as part of "intelligence" as being sped-up or made more efficient by quantum computation. I'd also be very interested in seeing citations for the claim that there are "many proposals for artificial general intelligence (AGI) using quantum computation."

WrongBot comments on Book Review: The Root of Thought - Less Wrong

Article Navigation

Comments (91)

Nearest Meetups

Virtual Study Room

Recent Comments

Recent Posts

Recent Rationality Quotes

Recent Wiki Edits

Top Contributors, 30 Days

Recent Karma Awards

Comments (91)

Top Contributors, 30 Days

You'll need to login or register to do that

(Don't worry, it only takes a few seconds)

Create

Login