[ExI] new neuron learning theory

Rafal Smigrodzki rafal.smigrodzki at gmail.com
Sat Apr 30 01:59:40 UTC 2022

On Fri, Apr 29, 2022 at 2:52 PM Darin Sunley via extropy-chat <
extropy-chat at lists.extropy.org> wrote:

> Perceptron-style "neurons" were a simplified caricature of how
> neurologists thought neurons /might/ work back in the 70s, even when they
> were first implemented.
> Time and neurological research hasn't been kind to the comparison.
> At this point, the only similarity between the basic elements of
> network-based machine learning algorithms and mammalian brain cells is the
> name. ML "neurons" are basically pure mathematical abstractions, completely
> unmoored from anything biological cells actually do.

### Biological neurons and ML neural nets that run on digital computers are
both physical implementations of the mathematical abstraction that, among
others, enables intelligence. So ML is not unmoored from biological cells,
it is a high-level abstract description of what the biological cells
actually do.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20220429/9db6aea1/attachment.htm>

More information about the extropy-chat mailing list