> I don't reject it, I just want to know the difference between saying "shit
> happens" and saying "it happened because of emergence". Yes, complicated
> systems behave in ways that are, well, complicated; but tell me something I
> didn't know.

The difference is that at some point people realised that it is
possible to create very simple models of neurons. One of the simplest
ones is a function that sums all of its inputs and then produces a
binary output according to weather a certain threshold is reached. You
can then introduce all sorts of details like connection weights,
non-binary triggers, spiking behaviours and so on.

Then you can think of algorithms that allow a network of artificial
neurons to self-organise. The simplest possible one might be Hebbian
learning: connection strengths are increased for neurons that fire
together.

There are some very nice models that combine artificial evolution with
artificial neural networks.

Artificial neural networks have been trained to fly planes, invest in
the stock market, converts speech to text, recognise handwriting and
so on and so on. For most of these cases, nobody understands how the
network works, they only understand how they created the necessary
conditions for a certain behaviour to emerge.

-- 
You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to everything-list+unsubscr...@googlegroups.com.
To post to this group, send email to everything-list@googlegroups.com.
Visit this group at http://groups.google.com/group/everything-list?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.


Reply via email to