On 12/17/2017 9:03 AM, Telmo Menezes wrote:
On Fri, Dec 8, 2017 at 7:53 PM, Brent Meeker <meeke...@verizon.net> wrote:
On 12/8/2017 2:24 AM, Telmo Menezes wrote:
On Thu, Dec 7, 2017 at 10:47 PM, Brent Meeker <meeke...@verizon.net>
wrote:
On 12/7/2017 1:01 AM, Telmo Menezes wrote:
On Wed, Dec 6, 2017 at 11:50 PM, Brent Meeker <meeke...@verizon.net>
wrote:
On 12/6/2017 1:46 AM, Bruno Marchal wrote:
I suspect that this is perhaps why Brent want to refer to the
environment
for relating consciousness to the machine, and in Artificial
Intelligence,
some people defend the idea that (mundane) consciousness occur only
when
the
environment contradicts a little bit the quasi automatic persistent
inference we do all the time.
That's Jeff Hawkins model of consciousness: one becomes conscious of
something when all lower, more specialized levels of the brain have
found
it
not to match their predictions.
In that sort of model, how does matter "know" that it is being used to
run a forecasting algorithm? Surely it doesn't right?
?? Why surely. It seems you're rejecting the idea that a physical system
can be conscious just out of prejudice.
Not at all. I remain agnostic on materialism vs. idealism. Maybe I am
even a strong agnostic: I suspect that the answer to this question
cannot be known.
Assuming materialism, consciousness must indeed be a property or
something that emerges from the interaction of fundamental particles,
the same way that, say, life does. Ok. All that I am saying is that
nobody has proposed any explanation of consciousness under this
assumption that I would call a theory. The above is not a theory, in
the same way that the Christian God is not a theory: it proposes to
explain a simple thing by appealing to a pre-existing more complex
thing -- in this case claiming that the act of forecasting at a very
high level somehow leads to consciousness, but without proposing any
first principles. It's a magical step.
What would a satisfactory (to you) first principle look like.
I cannot imagine one -- and this fuels my intuition that consciousness
is more fundamental than matter,
It fuels my intuition that it is a "wrong question".
and that emergentism is a dead-end.
But of course, my lack of imagination is not an argument. It could be
that I am too dumb/ignorant/crazy to come up with a good emergentist
theory. What I can -- and do -- is listen to any idea that comes up
and have an open mind. If you have one, I will gladly listen.
If we
consider the analogy of life, in the early 1900's when it was considered as
a chemical process all that could be said about it was that it involved
using energy to construct carbon based compounds and at a high level this
led to reproduction and natural selection and the origin of species. Now,
we have greatly elaborated on the molecular chemistry and can modify and
even created DNA and RNA molecules that realize "life". Where did we get
past the "magical step"? Or are you still waiting for "the atom of life" to
be discovered?
Here there is no magical step. Life can be understood all the way down
to basic chemistry. Ok, we don't have all the details, but we are not
missing anything fundamental. I am not waiting for the atoms of life
because I already know what they are. You just described them above.
Can you do that for consciousness?
Maybe not yet, but I can imagine what they might be: self-awareness,
construction of narratives about one's experiences, modeling other minds,...
What makes the hard problem hard is that it relates to a qualitatively
different phenomena than anything else that we try to understand. Life
can be talked about purely in the third person, but consciousness is
first person by definition.
So we are told. But what if someone could look at a recorded MRI of you
brain and tell you what you were thinking?
My view is that this sort of emergentism always smuggles a subtle but
important switcheroo at some point: moving from epistemology to
ontology.
For me, emergence is an epistemic tool. It is not possible for a human
to understand hyper-complex systems by considering all the variables
at the same time. We wouldn't be able to understand the human body
purely at the molecular level. So we create simplifying abstractions.
These abstractions have names such as "cells", "tissues", "organs",
"disease", etc etc. A Jupiter Brain might not need these tools. If
it's mind is orders of magnitude more complex than the human body,
then it could apprehend the entire thing at the molecular level, and
one could even say that this would lead to a higher level of
understanding than what we could hope for with our little monkey
brains.
In a sense, this would violate the very meaning of "understanding". If you
look at a website discussing the recent triumph of AlphaZero over Stockfish
in chess, there are arguments over whether the programs "understand" chess
or are they just very good at playing it. Those that claim the programs
don't understand chess mean that the programs just consult lots of memorized
positions and wether they led to a win or a loss. To "understand" chess
they should base their moves on some general principle which are simple
enough to explain to an amateur. In other words, to "understand" the game
is a social attribute = being able to explain it to a person. A lion knows
how to catch an antelope, but she doesn't understand it because she can't
explain it.
I would say that what we mean by "understanding" is having a model
(and I am going to repeat and agree with some things you say above),
that can:
- Make good predictions for behaviors under new conditions;
- Be communicable;
- Be constructive, in the sense that it can be combined to other
models about other things and fit nicely in a larger tapestry of
knowledge.
Humans can talk about cells and organs. Jupiter brains can talk about
swarms of molecules that require gigantic many-dimensional matrices to
describe. Emergence is a tool to overcome cognitive limits by creating
simpler levels of description. The magic step is to pretend that
creating a simpler level of description generates a new behavior.
If it succeeds in correctly predicting new behavior then as Aruthur C.
Clarke said, "Any sufficiently advanced technology is indistinguishable
from/magic/." and "indistinguishable" is a symmetric relation.
Brent
It
is literally "magical thinking" e.g. thinking that wearing a white
coat and a stethoscope makes you capable of diagnosing diseases.
In the case of biological systems, although we couldn't do what the
Jupiter Brain does, we could understand what are the first principles
that said brain would make use of.
Emergentists switch to the ontological. As if "emergence" generates
something new. As if it's something akin to a fundamental law of
nature. It's a language trick. When we say that something emerges from
something else, we are building an epistemic tool, we are not being
literal.
Is it a trick to say life emerges from chemistry?
No. It is an epistemic move. It is a trick to pretend that emergence
is ontological.
Could emergentism be true? Sure. But for it to be an actual theory, it
would have to provide some first principle. What I referred to as the
"atom" of consciousness, in the same way that a local trasaction
between two actors is the atom of economic models. Were is it?
The only way this could work is if the forecasting algorithm and the
cascading effects of failing predictions have the side effect of
creating the "right" sort of interactions at a lower level that
trigger consciousness.
In Hawkins model the predictions fail from the "bottom up", i.e. from the
subconscious, automatic responses up to the top/lanuage/conscious level.
I like Hawkins model and his work in general, but I think it is purely
about intelligence.
Then I want to know what these interactions
are, and what if the "atom" of consciousness, what is the first
principle. Without this, I would say that such hypothesis are not even
wrong.
There is no "atom of consciousness". In Hawkins model consciousness is
the
spreading to the 'failed prediction' signal across the top level of the
neocortex. As I said earlier, this is not Hawkins main interest, it's
more
an aside. He's more interested in intelligence.
Indeed. I read "On Intelligence" years ago (one decade?) so I might be
fuzzy on the details. I got the impression that he is completely
uninterested in consciousness, or that he doesn't even consider it a
serious question.
He does discuss it in the last chapter because he realizes it will be of
interest to the reader and that's where he speculates on an account similar
to Bruno's "waking up the boss".
John McCarthy was concerned that in creating an AI that we would
inadvertently create consciousness and thus incur ethical obligations we
were not prepared to meet.
Yes. I strongly agree with that concern.
Telmo.
Brent
But as has been discussed
here many times, philosophical zombies are probably not possible. That
would imply that a sufficiently intelligent system, however constructed,
will be conscious.
Yes, I tend to agree with this view.
Telmo.
Brent
Telmo.
Brent
--
You received this message because you are subscribed to the Google
Groups
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send
an
email to everything-list+unsubscr...@googlegroups.com.
To post to this group, send email to everything-list@googlegroups.com.
Visit this group at https://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an
email to everything-list+unsubscr...@googlegroups.com.
To post to this group, send email to everything-list@googlegroups.com.
Visit this group at https://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an
email to everything-list+unsubscr...@googlegroups.com.
To post to this group, send email to everything-list@googlegroups.com.
Visit this group at https://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to everything-list+unsubscr...@googlegroups.com.
To post to this group, send email to everything-list@googlegroups.com.
Visit this group at https://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
