Backpropagation is a supervised learning algorithm for multi-level neural networks. For each training example, it takes the difference between the expected and actual output at the final layer and then uses the differential of the sigmoid function at each node to work out error values at earlier layers and hence update the weights on links between nodes. It is an example of gradient descent.
Also known as: backprop
Used in glossary entries: gradient descent, neural network, sigmoid function, supervised learning