Why does Xavier initialization work?

Related questions

1. Can loss function negative?
2. What is dying ReLU?
3. What is weight initialization neural network?
4. What is Xavier initialization?
5. What is S shaped curve?
6. What is weight Ann?
7. What is J in machine learning?
8. Can I initialize a disk without losing data?
9. Can cost function be zero?
10. What is the formula for calculating cost function?

Tags

1. #networks
2. #weights
3. #neural
4. #other
5. #values
6. #xavier
7. #small
8. #motivation
9. #random
10. #network
11. #starting
12. #initialization
13. #functions
14. #large
15. #words
16. #neuron
17. #regions
18. #initialize
19. #saturated
20. #activation