Why does dropout regularization work?

Related questions

1. Which is better Lstm or GRU?
2. What is dropout in keras?
3. What is Timesteps in Lstm keras?
4. What is difference between CNN and RNN?
5. What is a GRU?
6. What is a good batch size?
7. Does batch size affect accuracy?
8. What are models in keras?
9. What is Lstm in keras?
10. What is stateful Lstm?

Tags

1. #thinning
2. #wider
3. #training
4. #overfitting
5. #layer
6. #nodes
7. #simple
8. #capacity
9. #dropout
10. #randomly
11. #network
12. #subsampled
13. #because
14. #outputs
15. #under
16. #required
17. #effect
18. #networks
19. #neural
20. #reducing
21. #during
22. #using
23. #prevent