Backpropagation is a training algorithm used for a multilayer neural networks, it allows for efficient computation of the gradient.
The backpropagation algorithm can be divided into several steps:
1) Forward propagation of training data through the network in order to generate output.
2) Use target value and output value to compute error derivative with respect to output.
3) Backpropagate to compute derivative of error with respect to output in the previous layer, and continue for all hidden layers.
4) Using derivatives with respect to error for output and all hidden layers to calculate the error derivative with respect to weights.
5) Update weights.