Re: nanowire/synases
- From: "N" <mimo_545@xxxxxxxxxxx>
- Date: 10 Sep 2006 09:09:15 -0700
sorting entropy in artificial axons.
Messing around with some graph paper and my flexi-curve
thinking about a spike train this pm.
<dink!>
Beginning with the idea that spike is an 'E' voltage
surge, on a hypothetical scale from A-E, with a
spike sine composed of accumulated 'noise' along an
pretend axon, at what point and any reason might
it be necessary to generate an (artificial) dendrite?.
might a dendrite be created for a containment, ie
primarily for the transmission of some voltage surges
away from the neuron, secondarily as a channel to a
synaptic logic gate? its form following the functional
necessity of the neuron to dissipate the excess energy?
and a synaptic 'gate' forming to cope to the maximum
(now fractional) mv from the cell.
<dok!>
What other information is already present in the
voltage/entropy chain that can be filtered out at a
dendrites junctions and at/through a synapse?
For example, a low voltage regular pulse may be just
as significant as high voltage irregular pulses to
convey data, but although I can understand how
energy dissipates, I can't see how any sine in a
flow is 'filtered' at a synaptic junction just yet,
(any suggestions?) although I've been thinking
around how neurotransmitters +/- charge as mean mV
spans could aid conduction of low energy sines.
<Zzztack!>
would artificial, pseudo-synapses set at mV
minimum and maximum invariables be sufficient
to transmit significants to adjacent neurons?
If a neuron will only fire at a certain peak, as
opposed to a succession of pulses in a train,
do all neurons fire at the same mV? or do
different sorts of neurons fire at different mVs?
<dok!>
Which spike trains are the strongest with the
highest mV? I guess ones about survival, guessing
lower energies might be used for peripherals.
Hmm,
<dink!>
N.
(news update
the wasp man came today...he said the nest was
2ft across :( I've got my windows shut for 4 hours)
<splat!!>
.
- Prev by Date: Re: Bayesian Inference Engine
- Next by Date: Re: Bayesian Inference Engine
- Previous by thread: Re: Bayesian Inference Engine
- Next by thread: Re: Gradual Learning, not Reinforcement Learning
- Index(es):
Relevant Pages
|
