Re: How do the brain neurons compute?

On 4 nov, 11:52, "Androcles" <Engin...@xxxxxxxxxxxxxxxx> wrote:
<s...@xxxxxxxxxxxx> wrote in message

news:1194191166.262878.56440@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
On 4 nov, 09:24, "Androcles" <Engin...@xxxxxxxxxxxxxxxx> wrote:



<s...@xxxxxxxxxxxx> wrote in message

news:1194183884.401753.9650@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
On 3 nov, 18:00, Tenifer <tensorsur...@xxxxxxxxx> wrote:

On Nov 4, 2:48 am, s...@xxxxxxxxxxxx wrote:

On 1 nov, 18:57, Tenifer <tensorsur...@xxxxxxxxx> wrote:

What kind of algorithm do our brain neurons use? How do they compute
to make us aware and produce our mind?

Neurologists chose their profession because they are poor in math.

Therefore only physicists excelling in mathematical algorithm can
figure out how the brain compute.

Anyone got an theory or mathematical idea how the brain neurons
compute (or what kind of algorithm it use)?

You should look up the work of Hebb in the 1940's. He completely
solved how neural networks correlate data. No algorithm required
except for simulating neural nets on linear computers.

The seat of awareness has long been established as being the
néocortex (a 6-layer neural network). You would have to dig into
neurophysiology archive to learn about this. Starting with
Pavlov, many others explored the various aspects. Look up
Eccles for references to most of them. Chauchard in the
1940's 1950's finished the job.

You could get hold of "On Intellignece" by Jeff Hawkins. He
rediscovered much of this and is good at explaining.

A pity that most in the pertaining disciplines don't even
know that all of this has been understood long ago.

No integrated refs in lectures for the past half century
ensured that these discoveries remained sleeping in
dusty archives.

André Michaud

So it is theoretically possible to build a self-aware computer
or android.

This also has been explored, and no. It will never be possible
because self-awareness requires life (sentience) to be involved
from the getgo so the network can chose by itself what is good
and what is bad for it.

Training real artificial neural nets with more than 1 layer to do
anything useful is extremely difficult since they can't be
programmed. On top of being technically difficult to build,
they have to be trained and guided for hours to be conditioned
into giving the correct answer or providing the correct reaction
for each set of input data.

I wonder when will we have these silicon sentient
beings joining with us in the planet. Do you know of folks
seriously building this now?

Some have tried and wasted their time. Can't be done.

André Michaud

It is theoretically possible to build a self-aware computer,
proven by "I think, therefore I am" -- Rene DesCartes.

Old Rene was sentient to start with before he could
work through the maze and get to that conclusion.

I for one won't be on alert in case one taps me on the
shoulder from behind as he utters "See ! Here I am.
You were wrong."

André Michaud

Nature has succeeded where man (a self-aware computer)
has (so far) failed. But man has only just begun trying, it took
Nature billions of years of trial and error. The issue isn't
whether man will succeed or not, but whether it is possible
to succeed. And it is. "Can't be done" applies to da Vinci's
helicopter, which cannot fly as designed. The underlying
principle behind it does fly, though.

In this case, "Can't be done" isn't based on any belief but
on the sum of consistant research done in the field.

You can build as complex an artificial neural net as
you want, it just won't have self awarenes from the
get go.

Self awareness is required to start with for any choice to
be made consistant with self survival. Artificial networks
have no preference. Unless extensively trained into a
specific direction for each possible choice they are
likely to have to make, they will just as readily chose
to turnoff their own power than to replace their battery.

They don't even don't care. They are inert.

I for one WILL be on alert in case one taps me on the
shoulder from behind as it utters "See ! Here I am.
You were right, André is asleep."

In your dreams, if you want my opinion.

Never underestimate the power of the human mind, André.
If I replaced a single neuron in my brain with an electronic
circuit that is FUNCTIONALLY identical and interfaces
with the rest (even if not physically identical), then I would
be unaware of the change.

Agreement. But you do have self awareness to start with.

Extend that concept to all my neurons and I can become
eternal. And a god.

.... Until the part that sustains your self awareness dies of
old age. then your eternal artificial neural net will stop
doing rational choices consistant with continued existance.

André Michaud

.

Relevant Pages

  • Re: How do the brain neurons compute?
    ... compute (or what kind of algorithm it use)? ... solved how neural networks correlate data. ... Never underestimate the power of the human mind, ... If I replaced a single neuron in my brain with an electronic ...
    (sci.physics)
  • Re: How do the brain neurons compute?
    ... figure out how the brain compute. ... compute (or what kind of algorithm it use)? ... solved how neural networks correlate data. ... except for simulating neural nets on linear computers. ...
    (sci.physics)
  • Re: How do the brain neurons compute?
    ... Therefore only physicists excelling in mathematical algorithm can ... Anyone got an theory or mathematical idea how the brain neurons ... scientists use for their neural networks. ...
    (sci.physics)
  • Re: How do the brain neurons compute?
    ... compute (or what kind of algorithm it use)? ... solved how neural networks correlate data. ... So it is theoretically possible to build a self-aware computer ... into giving the correct answer or providing the correct reaction ...
    (sci.physics)
  • Re: Artificial Intelligence and Neural Biology?
    ... programs with the neural knowledge to start a primitive brain. ... And as r norman has already posted, the neural networks we usually use ... are *extremely* crude imitations of real neurons. ... even our crude imitations have great power in principle. ...
    (talk.origins)