C.W.K.
Stream
Quiz · 4 questions

⛓️ Loss, Gradients, and Backpropagation

How neural networks actually learn

Level 0Curious
0 XP0/73 lessons0/11 achievements
0/120 XP to next level120 XP to go0% complete

Quiz

01What does it mean, mathematically, for a neural network to 'learn'?
02Which loss is the right default for multi-class classification with PyTorch?
03What is the most common practical fix for exploding gradients during training?
Hint
Vanishing and exploding gradients have different fixes — clipping is for the latter.
04Why is AdamW preferred over Adam-with-L2 for modern training?
Spotted a bug or have feedback on this page?Report an Issue

Comments 0

🔔 Reply notifications (sign in)
Sign inPlease sign in to comment.

No comments yet — be the first.