Thursday, December 25, 2008

A better algorithm

Algorithms based on causality are retro. I'm talking about "if...then." It is very time consuming to go through all the possibile outcomes in any given situation.

In a chess game, one cannot go through every possibility and analyze the outcomes because there are just too many combinations. A computer can do that but not a person. So experienced players will simply eliminate moves based on pattern recognition; i.e., they know that certain patterns in the chess position are not conducive to an advantage, so they eliminate them without looking into the specific sequence of moves that would result if a particular unfavorable move was played. They come up with a short list of candidate moves and then if necessary "think ahead" about the "if...then" ramifications.

Still, it would be more efficient to be able to eliminate all moves but one completely on the basis of pattern recognition. That could be very difficult to do. Ultimately, though, in life, it would be immensely freeing: to just know what the right thing to do, without having to think about it.

The mind, after all, knows nothing. It is a heuristic device, immensely useful, especially when employed Socratically to penetrate beneath the surface of appearances, but in terms of making choices, its algorithms are painfully slow and indeterminate. The heart and the gut are better choosers.

No comments: