4 ms·
It's not surprising that is the better solution. Humans do stuff the same way in real world. They repeat something until they get it right, and that's the learn
by mtgx 14y ago
It's not surprising that is the better solution. Humans do stuff the same way in real world. They repeat something until they get it right, and that's the learning process. The same with neural nets. You don't get to do something perfect from first try.