[ExI] Smallest human-equivalent device

Rafal Smigrodzki rafal.smigrodzki at gmail.com
Wed Oct 16 04:04:47 UTC 2013


On 10/11/13, Eugen Leitl <eugen at leitl.org> wrote:

> I agree. However, there's a widespread tendency to underestimate
> what evolutionary-driven biology has managed to accomplish in
> a few gigayears. A synapse is pretty damn small
>
> http://www.nature.com/nrn/journal/v12/n7/box/nrn3025_BX2.html
>
> Characteristic sizes of the synaptic complex
>
> Synaptic active zone diameter: 300 ± 150 nm
> Synaptic vesicle diameter: 35 ± 0.3 up to 50 nm
> Synaptic cleft width: 20 ± 2.8 nm
> Number of docked vesicles: 10 ± 5
> Total number of vesicle per synaptic bouton: 270 ± 180

### And not to forget the post-synaptic structures, which contain
hundreds of precisely tuned protein complexes that allow finely graded
ajustments of synaptic strength. Still, I am sure there is physics
that allows signal processing of the same type as synapse (adjustable
gain, summation, conditional gain adjustment) and does not use
diffusion and gets the job done in a smaller volume. Evolution is
locked in a wet organic substrate, designers of upload substrates are
not. Definitely many order of magnitude speed gains and most likely
substantial size reductions are possible here.

----------------

>
>> There should be large improvements just from removing metabolism from
>> the brain. Human brain already does a little bit of that: Neurons
>
> Metabolism is dual-use here, because the elements are active.
> You can complain about hydration, but diffusion is very efficient
> on microscale, and there *is* active transport. We can complain
> about homeostasis, but hardware configurability depends on
> the same mechanisms. In nanoscale solid-state you're pretty much
> limited to arrays of static elements with very low intrinsic
> connectivity, so you have to state the network layer in terms
> of such syntax. Look at 10^4 connectivity, and look at the
> space it takes.

### This is a very good remark. I do think that a neuromorphic system
could be done with spatially separated layers, a solid state gain
adjustment layer analogous to learning in existing synapses and a
network reconfiguration layer analogous to the synaptogenesis/synaptic
elimination aspect of learning. The reconfiguration layer may require
some molecular rather than electron movement but you should be still
able to keep metabolism out of most parts of the device.
--------------------
>
> 100-1 cm^3 vs. ~1400 cm^3 is a very large percentage.

### ??

Rafal




More information about the extropy-chat mailing list