Nota: A tradução desta entrada está atualmente em revisão de qualidade, portanto parte do conteúdo é exibida temporariamente apenas em inglês.
Este verbete ainda não foi traduzido para o seu idioma, portanto o original é exibido abaixo.
backpropagation
This term is a technical specialization within the field of machine learning and computational neuroscience. It describes a specific mathematical process of updating weights in a neural network to reduce error, making it a cornerstone of deep learning. Because it refers to a specific algorithmic method, it is used almost exclusively in academic, engineering, and data science contexts.
As a technical process or method, the word is treated as an uncountable noun. It describes the general mechanism of error propagation rather than a discrete object that can be counted. You would refer to the implementation of backpropagation or the process of backpropagation, rather than multiple backpropagations.
Meanings
Examples
The professor explained how backpropagation allows the network to learn from its mistakes.
I wonder if backpropagation is the most efficient way to handle this specific architecture.
We need to implement backpropagation to update the weights of the hidden layers.
Backpropagation is the cornerstone of modern deep learning.
Backpropagation is the cornerstone of modern deep learning.
Does this model rely on standard backpropagation or a modified version.
The gradient vanishes during backpropagation in very deep networks.
Let us look at the mathematical derivation of backpropagation.
Let us look at the mathematical derivation of backpropagation.
The system uses backpropagation to minimize the difference between predicted and actual values.