backpropagation

Terms from Artificial Intelligence: humans at the heart of algorithms

Page numbers are for draft copy at present; they will be replaced with correct numbers when final book is formatted. Chapter numbers are correct and will not change now.

Backpropagation is a supervised learning 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.

Used in Chap. 6: pages 84, 87; Chap. 7: page 105; Chap. 8: pages 111, 113; Chap. 9: pages 132, 133, 137, 139; Chap. 12: page 196; Chap. 20: page 344; Chap. 21: page 357

Also known as backprop