Synthesis

Feb 26, 2007 12:26

New idea. This one I like ( Read more... )

science/tech, artsy

Leave a comment

Comments 5

langostino February 26 2007, 18:52:05 UTC
. . . welcome to the world of evolutionary biology?

Reply

toastbuster February 26 2007, 19:41:34 UTC
Yeah, but self-conscious, and with flying cars and literary basis instead of natural selection.

Actually the iterative learning methods I was thinking about are called genetic algorithms.

Reply

langostino February 27 2007, 02:20:46 UTC
Actually the iterative learning methods I was thinking about are called genetic algorithms.

Isn't that a definition? I mean, "iterative"="generational", right?

Sorry I'm a downer. I just kind of see all processes as evolutionary--when you get down to it, natural selection is just the tautological statement "if it works, it works; if it don't, it don't."

But your idea for a learning synthesizer is awesome. I did some research in machine language learning; we should chat about it sometime.

Reply

toastbuster February 27 2007, 04:34:24 UTC
Aw, well sure. I agree with your definition. In computer science, thought, genetic algorithms are just a particular framework for iterative programs which imitate evolution very strictly--I was talking more about the convention than a strict definition.

But anyway, we should talk about it sometime. Also, I STILL have your housewarming present (I forgot to bring it on Saturday).

Reply


couchmonkee February 27 2007, 00:01:21 UTC
I made a comment, but LJ ate it. Here's a shorter version.
The learning synthesizer is crazy. Crazy awesome (if retardedly difficult to construct).
Same for the social simulation, although the contradictions in inputs could lead to some zany on-goings in sim-world.

Reply


Leave a comment

Up