From the course: Deep Learning: Getting Started
Join today to access over 25,900 courses taught by industry experts.
Back propagation
From the course: Deep Learning: Getting Started
Back propagation
Once we have estimated the prediction error from forward propagation, we need to do back propagation to adjust the weights and biases. What is the purpose of back propagation? Back propagation is used to understand how much each part of the neural network contributed to the overall error. During forward propagation, we calculate the overall error for the network. Each node in the neural network contributes to this error in some way. A node's contribution is influenced by the values of its weights and biases. Different nodes contribute differently depending on how well these weights and biases capture the relationship between the inputs and the target output. As we fine-tune the network, the weights and biases for each node needs to be adjusted to reduce its error contribution. Backpropagation helps calculate how much these values should be changed to lower the overall error. How does backpropagation work? Backpropagation works in the reverse direction of forward propagation moving…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.
