I believe there is a minor typo in the assignment. At the very end, it says:
“7.3.1 - Gradient Descent with Learning Rate Decay
Run the following code to see how the model does gradient descent and weight decay.” and I think it should be “and learning rate decay” instead of the weight. This same typo is in all three examples, so also 7.3.2 and 7.3.3.
1 Like
Thanks for bringing this up. The staff have been notified about this. Weight decay seems more related to regularization than learning rate schedule.
2 Likes