Geek, But...

I read a lot of different things. I'm not a Renaissance Man, by any means, but in certain broad categories, I like to pick up on whats interesting.

One of those areas is neural nets, which I know a little bit. Here's a very nice explanation from a NASA site: Neural net software learns by observing pairs of related patterns in the real world and learns to perform different tasks in response to different patterns: it discovers relationships by observation. A neural network consists of many interconnected processors (or nodes) using computational principles derived from neurons in the brain. Each node assigns a value to the input from each of its counterparts. As these values change, the network can adjust the way it responds. For example: if you wanted to automatically turn on your cars wipers, you might have one input that tracks whether there is wetness on the windshield; if the answer is yes, another might check whether the windshield washer was just used (if so, turn it on, briefly); while another might check how much wetness is on the windshield (a little - turn on 'intermittent'; a lot - turn on 'high'), and so forth.

People have been designing software to function like this for years, and its been pretty neat to read about. So when I read an article today about a bioscientist who was teaching an actual neural net, built up from neurons, not an Artificial Neural Net, which is done with software, I thought 'Well, thats interesting, but who needs to know that? I mean, its been done...' It took me a moment to realize 'Actually, no, it isn't. You're thinking of science fiction as reality again, Bill. Aka the Galactica Syndrome.'

Oh, yeah.