Seeing is believing; Machine vision; Teaching machines to see.(Science and Technology)(neural networks)

Economist (US), The, January, 2001

AS ANY parent will tell you, teaching a child how to tell the difference between types of object of the same class-animals, say, or letters of the alphabet-is a long job. A good way to speed it up is to point out specific, unique features. Squirrels have bushy tails. Zebra have stripes. Dogs have barks. And so on. Use the right feature and identification is straightforward. But how do you choose that feature in the first place? And what if you want to distinguish between things that are more abstract than animals?

These are questions that neural-network researchers have been grappling with for years. Neural networks are computer programs that are based loosely on animals' nervous systems. As such, they are able to learn from experience. But at the moment, the...

Read the rest of this article with a Free Trial at HighBeam Research.

Premium Content Partnership | HighBeam Research provides an in-depth online archive library of reference works.