Discussion Topic |
|
This thread has been locked |
WBraun
climber
|
|
|
Feb 14, 2019 - 03:00pm PT
|
No
It's actual meaning:
'Achintya' is a Sanskrit word meaning 'that which is beyond thought, imagination and contemplation'
|
|
eeyonkee
Trad climber
Golden, CO
|
|
|
Feb 14, 2019 - 04:15pm PT
|
As I had opined up thread, the fact that these neural networks cannot "explain" how they do what they do is a feature shared with the biological examples.
Nice insight, Ed! I would go further. I would opine that the answers from neural networks are part of what we mean by intelligence. The other part is an agent “in-charge” deciding what to tell the neural networks to focus on or how to interpret or direct the results.
The more I think of it, agency is the easy part. Organisms are the agents. Organisms are mainly just trying to stay alive and propagate. If they have evolved neural networks, then they will enlist these devices in their survival and propagation algorithms.
|
|
WBraun
climber
|
|
|
Feb 14, 2019 - 04:30pm PT
|
neural networks are part of what we mean by intelligence
Neural networks are the material hardware of the material organism (body).
Intelligence is far above that and originates from the soul the self itself .....
|
|
eeyonkee
Trad climber
Golden, CO
|
|
|
Feb 14, 2019 - 05:23pm PT
|
And another thing, neural networks are a great example of a "killer" design feature.
|
|
healyje
Trad climber
Portland, Oregon
|
|
|
Feb 14, 2019 - 05:32pm PT
|
Ed Hartouni wrote: As I had opined up thread, the fact that these neural networks cannot "explain" how they do what they do is a feature shared with the biological examples.
eeyonkee wrote: Nice insight, Ed! I would go further. I would opine that the answers from neural networks are part of what we mean by intelligence. The other part is an agent “in-charge” deciding what to tell the neural networks to focus on or how to interpret the results.
The more I think of it, agency is the easy part. Organisms are the agents. Organisms are mainly just trying to stay alive and propagate. If they have evolved neural networks, then they will enlist these devices in their survival and propagation algorithms.
I would disagree with both these assessments.
It is possible to understand how ML/DL/NN arrive at answers, but it's a complex/expensive thing to do and you have to understand both the architecture and algorithmic processions well enough to know where it's even worth looking because once they're churning it's a BIG space to be nosing around in, hence the google team's approach.
The only agency involved is the human programmers and so to me no intelligence, just algorithmic grinding.
|
|
eeyonkee
Trad climber
Golden, CO
|
|
|
Feb 14, 2019 - 05:54pm PT
|
So, are you suggesting a "ghost in the machine"? I mean what is so special about human agency that it could not be deconstructed into a hierarchical set of algorithms? I see this as a necessary condition of determinism. I really am interested in your take on this.
|
|
Ed Hartouni
Trad climber
Livermore, CA
|
|
|
Feb 14, 2019 - 06:52pm PT
|
It is possible to understand how ML/DL/NN arrive at answers, but it's a complex/expensive thing to do and you have to understand both the architecture and algorithmic processions well enough to know where it's even worth looking because once they're churning it's a BIG space to be nosing around in, hence the google team's approach.
maybe it's possible, but so far it's an open question (as far as I know)
nosing around in a 3 dimension space is bad enough, this is larger than that... usual methods will not work.
|
|
WBraun
climber
|
|
|
Feb 15, 2019 - 07:30am PT
|
There's no such thing as a "ghost in the machine".
Do you really think you are a ghost?
Modern scientists pride themselves so much in their logic, reasoning and thinking skills.
They seem very lacking according to their actual comments on these topics.
Extremely narrow minded, largely clueless and completely fixed on the machine, its hardware and oblivious to the operator itself ....
|
|
MH2
Boulder climber
Andy Cairns
|
|
|
Feb 15, 2019 - 08:09am PT
|
You could be describing yourself, Werner.
|
|
MikeL
Social climber
Southern Arizona
|
|
|
Feb 15, 2019 - 09:14am PT
|
There appears to be a confusion about the phrase. Ryle said the idea of dualistic entities (mind and body) by Descartes committed a category error by assuming they were in the same analytical domain. Koestler’s 1967 book that many of us might have read back in the day played on that idea, especially portraying (primitive) emotions as trumping or overpowering the logical intellect. These ideas are different and in different conversations than the common ideas of “ghosts.”
However, it might be safe to say that there is another notion caught by the term that most everyone has experience with. When it comes to explaining what’s going on in minds and in people’s behaviors, we just can’t say even if we had access to those minds objectively. Skinner finessed this issue in many of his writings by simply saying that mental operations did not matter; what mattered were behaviors. The mind was a black box.
The mind seems possessed by what we can’t completely or accurately say. Many things seem possessed in these ways—not just physical systems that we cannot completely account for, but also mental systems (like languages, cultures, societies, certain AI systems) that seem to have “minds” of their own. Everything seems to possess its own trajectory, and the interactions among trajectories lead to unresolvable complexity.
|
|
Ed Hartouni
Trad climber
Livermore, CA
|
|
|
Feb 15, 2019 - 10:09am PT
|
Everything seems to possess its own trajectory, and the interactions among trajectories lead to unresolvable complexity.
one way to look at it...
...if you take a look at a single atom, can you decide if that single atom is embedded in a gas, a liquid or a solid? That atom executes its own unique trajectory, it interacts with many other trajectories, and following 1e23 atoms is truly unresolvable.
Yet we know that there are materials we characterize as gas, liquid and solid, and these states of matter are describable, to some precision, in terms of measurable quantities, temperature, density, etc. Looking at a single atom you cannot tell the temperature of the collection of atoms nearby, nor the density of that collection.
But there is an order that is quite a bit simpler than the one that follows each individual atom around. And the ability to follow large numbers of atoms around, millions or even billions, provides a means of resolving their complexity providing a link to the simplicity.
|
|
Largo
Sport climber
The Big Wide Open Face
|
|
|
Topic Author's Reply - Feb 15, 2019 - 11:14am PT
|
So, are you suggesting a "ghost in the machine"? I mean what is so special about human agency that it could not be deconstructed into a hierarchical set of algorithms? I see this as a necessary condition of determinism.
------
This suggests that there is some classical "machine" all the way down, but all that's ever been shown at depth - from both a physicalist and observational POV is a field (for lack of a better word), and whatever arises from it has no stand-alone existence and is impermanent. Our "machines," and all forms, apparently arise from nothing then fade, as if ghosts.
Also, "determinism" as you use it is, I suspect, quite another thing from what occurs.
But the ghost in the machine metaphor is an interesting one to ponder. I especially like the thought experiment of a super duper, strictly objective machine from the future (which has a billion times the objectifying power of any present day machine) which hoves to on earth and conducts a comprehensive physical analysis of a human being.
Remember, the futuristic machine is strictly objective, meaning it has no experiential/subjective existence, isn't conscious that is is a machine conducting an investigation, and is just a fancy bucket of bolts and wires and diodes etc. running a futuristic algorithm.
The thought experiment is meant to intuitively show that no matter how exhaustive and comprehensive the machine's analysis, there would be no chance that it would ever suspect that the human was conscious because A) the machine, as a strictly objective mechanical object, has no analogue for nor experiential data stream from which it might possibly suspect let along detect consciousness, and B) there is nothing inherently present in the purely physical human form or design that would ever disclose or determine a conscious being, only a zombie just as experientially dead inside as the machine.
What's even more fascinating are the ham-fisted attempts to explain how the machine - having reached a level of complexity itself - WOULD be conscious, or that it could conclude the human was conscious through an objective analysis. Or that the thought experiment was itself misleading because it didn't take account of... cha cha cha.
What we perceive as machines (forms and objects) are ghosting in and out of existence in a space/time continuum that itself is an artifact of nothing at all, and which some report is consciousness itself.
What is "there" is not what we see, apparently.
Go figure...
|
|
Ed Hartouni
Trad climber
Livermore, CA
|
|
|
Feb 15, 2019 - 02:40pm PT
|
Or that the thought experiment was itself misleading because it didn't take account of...
the fundamental assumption that we understand our own subjective experience.
We do not.
We do experience it... we have no idea how that happens, the "thought experiment's" primary assumption is that something is happening that is different than what is happening in the machine.
What is it?
|
|
Largo
Sport climber
The Big Wide Open Face
|
|
|
Topic Author's Reply - Feb 15, 2019 - 03:35pm PT
|
the fundamental assumption that we understand our own subjective experience.
We do not.
We do experience it... we have no idea how that happens, the "thought experiment's" primary assumption is that something is happening that is different than what is happening in the machine.
You do not, Ed, because in my view, you believe that the "how" is a linear physical process, whereas others have said all the stuff, seemingly physical, the space in which this happens, the time in which it occurs, the location of the brain (and every "thing" else), and the apparent separateness between objects, to list a few, all issue FROM consciousness, which is nothing but the bubbling void.
By remaining close minded to any view but staunch physicalism, when you ask me to show you something, to explain something, you of course are asking for proof that conforms to a physicalist POV, since you have no other reference, and entertain no other possibilities, having no experience of same. Fair enough.
What's more, the point of the thought experiment is that the fundamental assumption that a purely objective analysis can disclose mind is mistaken. Meaning that the fundamental assumption of the thought experiment is NOT about an imagined physical understanding, rather that mind itself is not observable, measurable, or quantifiable. As I've said, the Hard Problem is a sucker's bet and a trick question because it assumes that a purely physical "explanation" of experience is possible (minus all the logical howlers like, "We only think it is different then brain function").
Then this: "Something is happening that is different than what is happening in the machine."
This strikes me (and Chalnmers et al) as magical or fantastic, that there should be any doubt that experience is "different" than dancing neurons. Logically speaking, this is a dead end. Physicalism can't brook something being more than it's parts, so even if you said that the brain "produces" or creates mind, you have more than dancing neurons (ie, experience). You might say experience isn't "real," because real and physical are selfsame. If you say the brain IS conscious, you have identity theory, and the insurmountable problems are obvious. Or you're left with an inherent quality that is neither observable or detectable. Remember that physicalism and emergence are incompatible. What emerges?
Also remember, even supervenience does not wash per mind and brain. Thanks to the thought experiments of Hilary Putnam (1973, 1975) and Tyler Burge (1979), it is widely held that intrinsic duplicates can indeed differ in the content of their mental states. Quite naturally, this is impossible with a machine model.
It is these and other concerns which have led progressive scientists to look at other ways to consider both brain and consciousness, beyond using a machine model. While Information Theory, biocentrisism and others are anything but worked out, they ope up what is otherwise a dead end, realizing as they do that future data and experiments cannot surmount the challenges just mentioned.
Consider this: The nature of physical stuff is deeply mysterious, and physics grows stranger by the hour. (Richard Feynman’s remark about quantum theory — “I think I can safely say that nobody understands quantum mechanics” — seems as true as ever.) Or rather, more carefully: The nature of physical stuff is mysterious except insofar as consciousness is itself a form of physical stuff.
This seems like a standard take of the staunch physicalist. What some of us are saying is that we might make some real progress (as is, no one has the slightest clue why we are conscious, the "why" defined as being a linear/causal/physical description) if we flip this wonky equation and consider that physical stuff is a form of consciousness.
Radical idea? Certainly.
|
|
WBraun
climber
|
|
|
Feb 15, 2019 - 03:55pm PT
|
Well ....
there goes the gross materialists neighborhood ..... :-)
|
|
Ed Hartouni
Trad climber
Livermore, CA
|
|
|
Feb 15, 2019 - 06:09pm PT
|
Physicalism can't brook something being more than it's parts...
that's probably not true in the way you mean it... where do the atoms in a "solid" have the notion of "solid"?
yet we have solids, with properties of solids, and apparently the individual atoms don't know anything about that...
but what I am getting at is your presumption that, in your thought experiments, the "mind" possesses attributes that you claim are not something a machine could have.
Not only have you avoided detailing what these attributes are, you have supported your presumption by invoking "common sense" and, very oddly, the peculiar (to you) behavior of quantum mechanics as they have been interpreted for you into the classical domain (which is where you sense stuff).
You have no experience regarding the quantum world, you only know what you read.
So your bottom line is that panpsychism is the model of consciousness.
However, this gives consciousness to everything. Why not a machine?
|
|
WBraun
climber
|
|
|
Feb 15, 2019 - 06:13pm PT
|
where does consciousness go when we are under anesthesia
Doesn't go anywhere, just can't feel the material body.
Sometimes the living entity leaves the material body and returns after the anesthesia wears off.
Paramatma takes care of the material body if you need to leave but ultimately return.
Many leave the body during dreaming and then return .....
|
|
zBrown
Ice climber
|
|
|
Feb 15, 2019 - 06:39pm PT
|
Getting in touch with some former associates about accessing some of this new rail hi tech guru software.
Could it be that DeepMind [sic] will soon be unleashed on the on the 2.5 million posts of the ST?
Can stoopidity be made totally predictable?
At what confidence level?
Oh mama, can this really be the end?
|
|
MH2
Boulder climber
Andy Cairns
|
|
|
Feb 15, 2019 - 07:17pm PT
|
Thank you, Dingus. Yes, general anesthesia is a mystery, as is what happens when you get knocked on the head and wake up with a similar "void" in your memory.
I don't like to say much about consciousness because I don't claim to know much about its (presumed) neurophysiology. I am content to have one and use it. As the podcast makes clear, surgeons and anesthesiologists are also pragmatists who happily go ahead and use what works without needing to know how it works.
It is quite surprising that such simple tests hadn't been done long ago. The podcast paints a pretty picture but isn't all that satisfactory as an explanation of how general anesthesia works.
It seems to me that what the podcast describes may be related to what Largo tells us about meditation. It could be that by quieting your own neural chatter you invoke the low frequency (1 cycle per second) spreading wave that disrupts the usual cross-talk between groups of neurons. If you could do to yourself what general anesthesia does, it might explain why meditators talk of emptiness and tell us that that is where the real deal is. It would be hard to report your state of mind during meditation if your mind was in the void.
I first made a reference to general anesthesia in the Creationists Take Another Called Strike thread:
http://www.supertopo.com/climbing/thread.php?topic_id=972999&tn=3061
The podcast also makes a case for starting an investigation of consciousness by looking at what happens under general anesthesia.
|
|
healyje
Trad climber
Portland, Oregon
|
|
|
Feb 15, 2019 - 07:21pm PT
|
Largo wrote: What some of us are saying is that we might make some real progress (as is, no one has the slightest clue why we are conscious, the "why" defined as being a linear/causal/physical description) if we flip this wonky equation and consider that physical stuff is a form of consciousness.
Jesus f*#king christ - it took you eight years of endlessly equivocating posts to finally just come out say that clearly in a single sentence? I mean, what the f*#k, you couldn't have just led with that in post #1 instead of #23,599? Crikey...
|
|
|
SuperTopo on the Web
|
Let us know!