Neural backpropagation

Neural backpropagation is a way for computers to learn things. When a computer or "neural network" is trying to learn something it starts by trying a guess and then changes its guess after it sees how close it was. It's kind of like when you are playing a game with your parents, like a board game or a card game. When you make a mistake, they tell you why it was wrong and how you can change it to make it right. The same thing happens when a computer is learning with neural backpropagation. It might guess something wrong, but then it can go back (called "backpropagation") and fix it by making a new guess that is closer to being right. It can do this a lot of times, until it finally guesses all of the answers correctly.