1. A machine learning algorithm used to train artificial neural networks by calculating the gradient of the loss function with respect to each weight by the chain rule, propagating the error backwards through the network layers.
Error backpropagation is the fundamental algorithm used in training deep neural networks.
A retropropagação de erro é o algoritmo fundamental usado no treinamento de redes neurais profundas.
2. The process of computing partial derivatives of the error function through the network in reverse order, from output layer to input layer.
During error backpropagation, gradients flow backwards through the network to update the weights.
Durante a retropropagação de erro, os gradientes fluem para trás pela rede para atualizar os pesos.