[ExI] ANN question

Adrian Tymes atymes at gmail.com
Thu Mar 16 17:57:14 UTC 2017

On Wed, Mar 15, 2017 at 1:40 PM, William Flynn Wallace
<foozler83 at gmail.com> wrote:
> On Wed, Mar 15, 2017 at 3:09 PM, William Flynn Wallace <foozler83 at gmail.com>
> wrote:
>> OK, I think we are in agreement:  you and the ANN at looking at the state
>> of the neuron at a fixed point in time, and I am looking at it over a period
>> of time.  Maybe we need to look at both to understand the neuron

Pretty much - though I'm just explaining what the ANNs' simplification
(relative to what you're thinking) is.  What you're looking at is
closer to understanding actual neurons.

> addition - come to think of it, whether the neuron is firing or not at a
> specific point in time, is a function of where on its body you measure the
> chemical exchange;  we know that the spike travels the length of the cell
> body, a traveling wave, such that at the receiving end - the dendrite - you
> can measure the influx of ions into the body, whereas at its opposite end
> nothing is happening yet.  Maybe this makes no difference.  What do I know?

Oh, it can make a difference.  Perhaps it would be better to say that
ANNs model synapses and the things directly connected to the synapse
(which is only parts of the neurons on each side), rather than entire
neurons (which might be considered to have multiple
if-x-then-input-to-next-stage bits inside themselves).  This might be
akin to the difference between an individual transistor or gate, and
an entire chip (neuron) that is itself part of a larger circuit board

More information about the extropy-chat mailing list