https://www.coursera.org/learn/advanced-learning-algorithms/lecture/rhcTZ/computation-graph-optional

Course Q&A Machine Learning Specialization Advanced Learning Algorithms
1

When creating a post, please add:

  • Week # must be added in the tags option of the post.
  • Link to the classroom item you are referring to:
  • Description (include relevant info but please do not post solution code or your entire notebook):

image
image1182×530 77.2 KB

I cannot understand why the number of steps is N + P can anyone please tell all the nodes and show all the calculations showing it’s step no ?
So that it will be clear that it is in fact N + P.

Regards
Vedant Sharma

2

Hi @Vedbhai

In backward propagation, each node’s activation gradient is computed once, giving N steps and each parameter’s gradient is computed once, giving P steps. Therefore, the total computational complexity is N + P.

If you need more help, feel free to ask, and I can explain it with an example.

1 Like