From: Stathis Papaioannou (stathisp@gmail.com)
Date: Fri Nov 16 2007 - 05:56:14 MST
On 16/11/2007, Adam Safron <asafron@gmail.com> wrote:
> If you're modeling based on functional connections between neurons,
> this does not imply that you need to understand how the neuronal level
> leads to higher-level cognitive functions.
I find it hard to believe that full neuronal emulation would be a
lesser technical challenge than understanding higher level cognitive
functions. However, even if emulation comes first, it would provide a
research tool far superior to anything we had before in neuroscience.
You could tweak a few parameters, rewire a few connections, add or
subtract neuronal components all in software and observe the effect.
My original point to Wei Dai was that if we could simulate neural
tissue doing something "simple" at a fine enough level, it would be a
matter of scaling to simulate neural tissue doing something "complex",
since the neurons involved in abstract thought don't contain special
non-computable parts that are lacking in more pedestrian neurons.
> However, a properly
> functioning neuron-by-neuron model of the brain would probably not
> require molecular-level resolution.
That is often assumed in discussions about mind uploading, but I
haven't seen any good justification for it. Tiny concentrations of
small molecules like LSD non-covalently binding to neuronal proteins
and slightly changing their conformation can cause drastic changes in
cognition. Millions of other parameters per neuron, including its
shape, its relationship to all its neighbours, the pattern of surface
receptors, the state of cytoplasmic biochemical activity involved in
making and removing neurotransmitters and receptors, and so on, are
all important even if you are just the considering a binary question
such as whether the neuron will or won't fire. How are you going to
model the behaviour of the neuron in a brain without considering
detail at this level?
> However, it is unclear that this will constitute a very plausible
> means of creating a super-intelligence. A brain-in-a-box would not
> know how to improve its own functioning. Even if you model our best
> minds down to the molecular level, no one alive today would know how
> to improve the intelligence via neural modifications (besides making
> everything uniformly faster). The brain is a non-linear, self-
> organizing system. Emulation is not understanding.
But it would enormously enhance the process of understanding. I
imagine that even crude trial and error experimentation with the
nascent software would quickly result in cognitive improvements. (The
ethical considerations in undertaking such experiments are a different
matter.)
> If you use a non-molecular modeling approach, you will not start with
> a fully functioning sentience. You will have an AI-child that you
> may be able to nurture into a some modicum of sentient functioning.
> There is a good possibility that it would be insane, as human neural
> organization co-arises with embodied experience in the world. Also,
> children do not start out with fully-developed brains. You would have
> to have a detailed model of how neurodevelopment and neuroplasticity
> work over time. We'll get to this state of understanding one day, but
> even then, it's not clear that this is a good method for achieving
> super-intelligence.
I'm guessing that a model above the molecular level just won't work at
all, like a half-baked computer chip won't work at all. I expect that
higher level modelling of brain function will come far earlier than
true emulation, and human-level AI will come before AI modelled on the
human brain. This has generally been the pattern in every other
endeavour where humans try to copy a biological system. Wheels come
before walking robots, and walking robots come before androids with
artificial muscle, bone, circulatory and nervous tissue.
> But this is all assuming that we're talking about emulation. Assuming
> advanced nano-technology and advanced knowledge in molecular/cellular
> neurobiology (not molecule-by-molecule modeling capabilities) here's
> an idea which I have been toying with lately: 1) Flood a person's
> brain with nanobots. 2) Have the nanobots determine the functional
> properties of every neuron and glial cell in the brain (including
> endocrine functions). 3) Have the nanobots replace each of the cells
> with a functionally-identical synthetic equivalent. 4) Connect the
> synthetic brain to an artificial body or virtual body. If you do
> these things, we will have been limited from many of the constraints
> of our biology. In addition to overcoming the fact that our brains
> age, we could probably speed up cognition by several orders of
> magnitude (if your embodiment is in a virtual world that could keep-
> up). But the time-scales of developing these technologies may make
> this idea irrelevant. By the time we have these capabilities, we may
> have already developed an AGI that figured out a better way.
Yes.
-- Stathis Papaioannou
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:01:00 MDT