# optimization – ok, that makes sense. I don’t usually think of optimization as re-programming

optimization – ok, that makes sense. I don’t usually think of optimization as re-programming although they can be modeled in a program. Map and Reduce, least-squares, shortest-path etc are the kinds of things that ‘nature’ seems to do quite well all by itself, but for us to model them has proven to be challenging but we’ve had some great successes so far.

[ie – trying to find an algorithm for crumbling up a piece of paper into a ball is a _very_ difficult mathematical problem, but for us, just take a piece of paper and make a ball out of it]

But I can see how it’s considered reprogramming as well.

My favorite is just find diagonal (furthest distance) and half it, then start making triangles out of that, solving at the point where you have a manageable triangle to work with — a kind of divide and conquer. I don’t know what that’s called – probably some kind of least-squares maybe… but it’s a method I use real-world when I want to conquer a large space into manageable parts.

=====